diff --git a/backend/api/blog_writer/router.py b/backend/api/blog_writer/router.py
index 0d3b60fe..12c6a72e 100644
--- a/backend/api/blog_writer/router.py
+++ b/backend/api/blog_writer/router.py
@@ -185,10 +185,20 @@ async def get_research_status(task_id: str) -> Dict[str, Any]:
 
 # Outline Endpoints
 @router.post("/outline/start")
-async def start_outline_generation(request: BlogOutlineRequest) -> Dict[str, Any]:
+async def start_outline_generation(
+    request: BlogOutlineRequest,
+    current_user: Dict[str, Any] = Depends(get_current_user)
+) -> Dict[str, Any]:
     """Start an outline generation operation and return a task ID for polling."""
     try:
-        task_id = task_manager.start_outline_task(request)
+        # Extract Clerk user ID (required)
+        if not current_user:
+            raise HTTPException(status_code=401, detail="Authentication required")
+        user_id = str(current_user.get('id'))
+        if not user_id:
+            raise HTTPException(status_code=401, detail="User ID not found in authentication token")
+        
+        task_id = task_manager.start_outline_task(request, user_id)
         return {"task_id": task_id, "status": "started"}
     except Exception as e:
         logger.error(f"Failed to start outline generation: {e}")
@@ -272,12 +282,22 @@ async def generate_section(request: BlogSectionRequest) -> BlogSectionResponse:
 
 
 @router.post("/content/start")
-async def start_content_generation(request: Dict[str, Any]) -> Dict[str, Any]:
+async def start_content_generation(
+    request: Dict[str, Any],
+    current_user: Dict[str, Any] = Depends(get_current_user)
+) -> Dict[str, Any]:
     """Start full content generation and return a task id for polling.
 
     Accepts a payload compatible with MediumBlogGenerateRequest to minimize duplication.
     """
     try:
+        # Extract Clerk user ID (required)
+        if not current_user:
+            raise HTTPException(status_code=401, detail="Authentication required")
+        user_id = str(current_user.get('id'))
+        if not user_id:
+            raise HTTPException(status_code=401, detail="User ID not found in authentication token")
+        
         # Map dict to MediumBlogGenerateRequest for reuse
         from models.blog_models import MediumBlogGenerateRequest, MediumSectionOutline, PersonaInfo
         sections = [MediumSectionOutline(**s) for s in request.get("sections", [])]
@@ -293,7 +313,7 @@ async def start_content_generation(request: Dict[str, Any]) -> Dict[str, Any]:
             globalTargetWords=request.get("globalTargetWords", 1000),
             researchKeywords=request.get("researchKeywords") or request.get("keywords"),
         )
-        task_id = task_manager.start_content_generation_task(req)
+        task_id = task_manager.start_content_generation_task(req, user_id)
         return {"task_id": task_id, "status": "started"}
     except Exception as e:
         logger.error(f"Failed to start content generation: {e}")
@@ -307,6 +327,51 @@ async def content_generation_status(task_id: str) -> Dict[str, Any]:
         status = await task_manager.get_task_status(task_id)
         if status is None:
             raise HTTPException(status_code=404, detail="Task not found")
+        
+        # If task failed with subscription error, return HTTP error so frontend interceptor can catch it
+        if status.get('status') == 'failed' and status.get('error_status') in [429, 402]:
+            error_data = status.get('error_data', {}) or {}
+            error_status = status.get('error_status', 429)
+            
+            if not isinstance(error_data, dict):
+                logger.warning(f"Content generation task {task_id} error_data not dict: {error_data}")
+                error_data = {'error': str(error_data)}
+            
+            # Determine provider and usage info
+            stored_error_message = status.get('error', error_data.get('error'))
+            provider = error_data.get('provider', 'unknown')
+            usage_info = error_data.get('usage_info')
+            
+            if not usage_info:
+                usage_info = {
+                    'provider': provider,
+                    'message': stored_error_message,
+                    'error_type': error_data.get('error_type', 'unknown')
+                }
+                # Include any known fields from error_data
+                for key in ['current_tokens', 'requested_tokens', 'limit', 'current_calls']:
+                    if key in error_data:
+                        usage_info[key] = error_data[key]
+            
+            # Build error message for detail
+            error_msg = error_data.get('message', stored_error_message or 'Subscription limit exceeded')
+            
+            # Log the subscription error with all context
+            logger.warning(f"Content generation task {task_id} failed with subscription error {error_status}: {error_msg}")
+            logger.warning(f"   Provider: {provider}, Usage Info: {usage_info}")
+            
+            # Use JSONResponse to ensure detail is returned as-is, not wrapped in an array
+            from fastapi.responses import JSONResponse
+            return JSONResponse(
+                status_code=error_status,
+                content={
+                    'error': error_data.get('error', stored_error_message or 'Subscription limit exceeded'),
+                    'message': error_msg,
+                    'provider': provider,
+                    'usage_info': usage_info
+                }
+            )
+        
         return status
     except HTTPException:
         raise
@@ -499,14 +564,24 @@ async def get_outline_cache_entries(limit: int = 20):
 # ---------------------------
 
 @router.post("/generate/medium/start")
-async def start_medium_generation(request: MediumBlogGenerateRequest):
+async def start_medium_generation(
+    request: MediumBlogGenerateRequest,
+    current_user: Dict[str, Any] = Depends(get_current_user)
+):
     """Start medium-length blog generation (≤1000 words) and return a task id."""
     try:
+        # Extract Clerk user ID (required)
+        if not current_user:
+            raise HTTPException(status_code=401, detail="Authentication required")
+        user_id = str(current_user.get('id'))
+        if not user_id:
+            raise HTTPException(status_code=401, detail="User ID not found in authentication token")
+        
         # Simple server-side guard
         if (request.globalTargetWords or 1000) > 1000:
             raise HTTPException(status_code=400, detail="Global target words exceed 1000; use per-section generation")
 
-        task_id = task_manager.start_medium_generation_task(request)
+        task_id = task_manager.start_medium_generation_task(request, user_id)
         return {"task_id": task_id, "status": "started"}
     except HTTPException:
         raise
@@ -522,6 +597,51 @@ async def medium_generation_status(task_id: str):
         status = await task_manager.get_task_status(task_id)
         if status is None:
             raise HTTPException(status_code=404, detail="Task not found")
+        
+        # If task failed with subscription error, return HTTP error so frontend interceptor can catch it
+        if status.get('status') == 'failed' and status.get('error_status') in [429, 402]:
+            error_data = status.get('error_data', {}) or {}
+            error_status = status.get('error_status', 429)
+            
+            if not isinstance(error_data, dict):
+                logger.warning(f"Medium generation task {task_id} error_data not dict: {error_data}")
+                error_data = {'error': str(error_data)}
+            
+            # Determine provider and usage info
+            stored_error_message = status.get('error', error_data.get('error'))
+            provider = error_data.get('provider', 'unknown')
+            usage_info = error_data.get('usage_info')
+            
+            if not usage_info:
+                usage_info = {
+                    'provider': provider,
+                    'message': stored_error_message,
+                    'error_type': error_data.get('error_type', 'unknown')
+                }
+                # Include any known fields from error_data
+                for key in ['current_tokens', 'requested_tokens', 'limit', 'current_calls']:
+                    if key in error_data:
+                        usage_info[key] = error_data[key]
+            
+            # Build error message for detail
+            error_msg = error_data.get('message', stored_error_message or 'Subscription limit exceeded')
+            
+            # Log the subscription error with all context
+            logger.warning(f"Medium generation task {task_id} failed with subscription error {error_status}: {error_msg}")
+            logger.warning(f"   Provider: {provider}, Usage Info: {usage_info}")
+            
+            # Use JSONResponse to ensure detail is returned as-is, not wrapped in an array
+            from fastapi.responses import JSONResponse
+            return JSONResponse(
+                status_code=error_status,
+                content={
+                    'error': error_data.get('error', stored_error_message or 'Subscription limit exceeded'),
+                    'message': error_msg,
+                    'provider': provider,
+                    'usage_info': usage_info
+                }
+            )
+        
         return status
     except HTTPException:
         raise
diff --git a/backend/api/blog_writer/seo_analysis.py b/backend/api/blog_writer/seo_analysis.py
index dee0df5e..84deaf21 100644
--- a/backend/api/blog_writer/seo_analysis.py
+++ b/backend/api/blog_writer/seo_analysis.py
@@ -5,7 +5,7 @@ Provides API endpoint for analyzing blog content SEO with parallel processing
 and CopilotKit integration for real-time progress updates.
 """
 
-from fastapi import APIRouter, HTTPException, BackgroundTasks
+from fastapi import APIRouter, HTTPException, BackgroundTasks, Depends
 from pydantic import BaseModel
 from typing import Dict, Any, Optional
 from loguru import logger
@@ -13,6 +13,7 @@ from datetime import datetime
 
 from services.blog_writer.seo.blog_content_seo_analyzer import BlogContentSEOAnalyzer
 from services.blog_writer.core.blog_writer_service import BlogWriterService
+from middleware.auth_middleware import get_current_user
 
 
 router = APIRouter(prefix="/api/blog-writer/seo", tags=["Blog SEO Analysis"])
@@ -56,7 +57,10 @@ blog_writer_service = BlogWriterService()
 
 
 @router.post("/analyze", response_model=SEOAnalysisResponse)
-async def analyze_blog_seo(request: SEOAnalysisRequest):
+async def analyze_blog_seo(
+    request: SEOAnalysisRequest,
+    current_user: Dict[str, Any] = Depends(get_current_user)
+):
     """
     Analyze blog content for SEO optimization
     
@@ -69,6 +73,7 @@ async def analyze_blog_seo(request: SEOAnalysisRequest):
     
     Args:
         request: SEOAnalysisRequest containing blog content and research data
+        current_user: Authenticated user from middleware
         
     Returns:
         SEOAnalysisResponse with comprehensive analysis results
@@ -76,6 +81,14 @@ async def analyze_blog_seo(request: SEOAnalysisRequest):
     try:
         logger.info(f"Starting SEO analysis for blog content")
         
+        # Extract Clerk user ID (required)
+        if not current_user:
+            raise HTTPException(status_code=401, detail="Authentication required")
+        
+        user_id = str(current_user.get('id', ''))
+        if not user_id:
+            raise HTTPException(status_code=401, detail="Invalid user ID in authentication token")
+        
         # Validate request
         if not request.blog_content or not request.blog_content.strip():
             raise HTTPException(status_code=400, detail="Blog content is required")
@@ -91,7 +104,8 @@ async def analyze_blog_seo(request: SEOAnalysisRequest):
         analysis_results = await seo_analyzer.analyze_blog_content(
             blog_content=request.blog_content,
             research_data=request.research_data,
-            blog_title=request.blog_title
+            blog_title=request.blog_title,
+            user_id=user_id
         )
         
         # Check for errors
@@ -131,7 +145,10 @@ async def analyze_blog_seo(request: SEOAnalysisRequest):
 
 
 @router.post("/analyze-with-progress")
-async def analyze_blog_seo_with_progress(request: SEOAnalysisRequest):
+async def analyze_blog_seo_with_progress(
+    request: SEOAnalysisRequest,
+    current_user: Dict[str, Any] = Depends(get_current_user)
+):
     """
     Analyze blog content for SEO with real-time progress updates
     
@@ -140,6 +157,7 @@ async def analyze_blog_seo_with_progress(request: SEOAnalysisRequest):
     
     Args:
         request: SEOAnalysisRequest containing blog content and research data
+        current_user: Authenticated user from middleware
         
     Returns:
         Generator yielding progress updates and final results
@@ -147,6 +165,14 @@ async def analyze_blog_seo_with_progress(request: SEOAnalysisRequest):
     try:
         logger.info(f"Starting SEO analysis with progress for blog content")
         
+        # Extract Clerk user ID (required)
+        if not current_user:
+            raise HTTPException(status_code=401, detail="Authentication required")
+        
+        user_id = str(current_user.get('id', ''))
+        if not user_id:
+            raise HTTPException(status_code=401, detail="Invalid user ID in authentication token")
+        
         # Validate request
         if not request.blog_content or not request.blog_content.strip():
             raise HTTPException(status_code=400, detail="Blog content is required")
@@ -209,7 +235,9 @@ async def analyze_blog_seo_with_progress(request: SEOAnalysisRequest):
                 # Perform actual analysis
                 analysis_results = await seo_analyzer.analyze_blog_content(
                     blog_content=request.blog_content,
-                    research_data=request.research_data
+                    research_data=request.research_data,
+                    blog_title=request.blog_title,
+                    user_id=user_id
                 )
                 
                 # Final result
diff --git a/backend/api/blog_writer/task_manager.py b/backend/api/blog_writer/task_manager.py
index acbc15f0..59952a18 100644
--- a/backend/api/blog_writer/task_manager.py
+++ b/backend/api/blog_writer/task_manager.py
@@ -88,8 +88,12 @@ class TaskManager:
                 response["error"] = task["error"]
                 if "error_status" in task:
                     response["error_status"] = task["error_status"]
+                    logger.info(f"[TaskManager] get_task_status for {task_id}: Including error_status={task['error_status']} in response")
                 if "error_data" in task:
                     response["error_data"] = task["error_data"]
+                    logger.info(f"[TaskManager] get_task_status for {task_id}: Including error_data with keys: {list(task['error_data'].keys()) if isinstance(task['error_data'], dict) else 'not-dict'}")
+                else:
+                    logger.warning(f"[TaskManager] get_task_status for {task_id}: Task failed but no error_data found. Task keys: {list(task.keys())}")
             
             return response
     
@@ -127,29 +131,33 @@ class TaskManager:
             asyncio.create_task(self._run_research_task(task_id, request, user_id))
             return task_id
     
-    def start_outline_task(self, request: BlogOutlineRequest) -> str:
+    def start_outline_task(self, request: BlogOutlineRequest, user_id: str) -> str:
         """Start an outline generation operation and return a task ID."""
         task_id = self.create_task("outline")
         
         # Start the outline generation operation in the background
-        asyncio.create_task(self._run_outline_generation_task(task_id, request))
+        asyncio.create_task(self._run_outline_generation_task(task_id, request, user_id))
         
         return task_id
 
-    def start_medium_generation_task(self, request: MediumBlogGenerateRequest) -> str:
+    def start_medium_generation_task(self, request: MediumBlogGenerateRequest, user_id: str) -> str:
         """Start a medium (≤1000 words) full-blog generation task."""
         task_id = self.create_task("medium_generation")
-        asyncio.create_task(self._run_medium_generation_task(task_id, request))
+        asyncio.create_task(self._run_medium_generation_task(task_id, request, user_id))
         return task_id
 
-    def start_content_generation_task(self, request: MediumBlogGenerateRequest) -> str:
+    def start_content_generation_task(self, request: MediumBlogGenerateRequest, user_id: str) -> str:
         """Start content generation (full blog via sections) with provider parity.
 
         Internally reuses medium generator pipeline for now but tracked under
         distinct task_type 'content_generation' and same polling contract.
+        
+        Args:
+            request: Content generation request
+            user_id: User ID (required for subscription checks and usage tracking)
         """
         task_id = self.create_task("content_generation")
-        asyncio.create_task(self._run_medium_generation_task(task_id, request))
+        asyncio.create_task(self._run_medium_generation_task(task_id, request, user_id))
         return task_id
     
     async def _run_research_task(self, task_id: str, request: BlogResearchRequest, user_id: str):
@@ -205,7 +213,7 @@ class TaskManager:
                     self.task_storage[task_id]["status"] = "failed"
                     self.task_storage[task_id]["error"] = "Research completed with unknown status"
     
-    async def _run_outline_generation_task(self, task_id: str, request: BlogOutlineRequest):
+    async def _run_outline_generation_task(self, task_id: str, request: BlogOutlineRequest, user_id: str):
         """Background task to run outline generation and update status with progress messages."""
         try:
             # Update status to running
@@ -215,21 +223,31 @@ class TaskManager:
             # Send initial progress message
             await self.update_progress(task_id, "🧩 Starting outline generation...")
             
-            # Run the actual outline generation with progress updates
-            result = await self.service.generate_outline_with_progress(request, task_id)
+            # Run the actual outline generation with progress updates (pass user_id for subscription checks)
+            result = await self.service.generate_outline_with_progress(request, task_id, user_id)
             
             # Update status to completed
             await self.update_progress(task_id, f"✅ Outline generated successfully! Created {len(result.outline)} sections with {len(result.title_options)} title options.")
             self.task_storage[task_id]["status"] = "completed"
             self.task_storage[task_id]["result"] = result.dict()
             
+        except HTTPException as http_error:
+            # Handle HTTPException (e.g., 429 subscription limit) - preserve error details for frontend
+            error_detail = http_error.detail
+            error_message = error_detail.get('message', str(error_detail)) if isinstance(error_detail, dict) else str(error_detail)
+            await self.update_progress(task_id, f"❌ {error_message}")
+            self.task_storage[task_id]["status"] = "failed"
+            self.task_storage[task_id]["error"] = error_message
+            # Store HTTP error details for frontend modal
+            self.task_storage[task_id]["error_status"] = http_error.status_code
+            self.task_storage[task_id]["error_data"] = error_detail if isinstance(error_detail, dict) else {"error": str(error_detail)}
         except Exception as e:
             await self.update_progress(task_id, f"❌ Outline generation failed: {str(e)}")
             # Update status to failed
             self.task_storage[task_id]["status"] = "failed"
             self.task_storage[task_id]["error"] = str(e)
 
-    async def _run_medium_generation_task(self, task_id: str, request: MediumBlogGenerateRequest):
+    async def _run_medium_generation_task(self, task_id: str, request: MediumBlogGenerateRequest, user_id: str):
         """Background task to generate a medium blog using a single structured JSON call."""
         try:
             self.task_storage[task_id]["status"] = "running"
@@ -245,6 +263,7 @@ class TaskManager:
             result: MediumBlogGenerateResult = await self.service.generate_medium_blog_with_progress(
                 request,
                 task_id,
+                user_id
             )
 
             if not result or not getattr(result, "sections", None):
@@ -263,10 +282,38 @@ class TaskManager:
             self.task_storage[task_id]["result"] = result.dict()
             await self.update_progress(task_id, f"✅ Generated {len(result.sections)} sections successfully.")
 
-        except Exception as e:
-            await self.update_progress(task_id, f"❌ Medium generation failed: {str(e)}")
+        except HTTPException as http_error:
+            # Handle HTTPException (e.g., 429 subscription limit) - preserve error details for frontend
+            logger.info(f"[TaskManager] Caught HTTPException in medium generation task {task_id}: status={http_error.status_code}, detail={http_error.detail}")
+            error_detail = http_error.detail
+            error_message = error_detail.get('message', str(error_detail)) if isinstance(error_detail, dict) else str(error_detail)
+            await self.update_progress(task_id, f"❌ {error_message}")
             self.task_storage[task_id]["status"] = "failed"
-            self.task_storage[task_id]["error"] = str(e)
+            self.task_storage[task_id]["error"] = error_message
+            # Store HTTP error details for frontend modal
+            self.task_storage[task_id]["error_status"] = http_error.status_code
+            self.task_storage[task_id]["error_data"] = error_detail if isinstance(error_detail, dict) else {"error": str(error_detail)}
+            logger.info(f"[TaskManager] Stored error_status={http_error.status_code} and error_data keys: {list(error_detail.keys()) if isinstance(error_detail, dict) else 'not-dict'}")
+        except Exception as e:
+            # Check if this is an HTTPException that got wrapped (can happen in async tasks)
+            # HTTPException has status_code and detail attributes
+            logger.info(f"[TaskManager] Caught Exception in medium generation task {task_id}: type={type(e).__name__}, has_status_code={hasattr(e, 'status_code')}, has_detail={hasattr(e, 'detail')}")
+            if hasattr(e, 'status_code') and hasattr(e, 'detail'):
+                # This is an HTTPException that was caught as generic Exception
+                logger.info(f"[TaskManager] Detected HTTPException in Exception handler: status={e.status_code}, detail={e.detail}")
+                error_detail = e.detail
+                error_message = error_detail.get('message', str(error_detail)) if isinstance(error_detail, dict) else str(error_detail)
+                await self.update_progress(task_id, f"❌ {error_message}")
+                self.task_storage[task_id]["status"] = "failed"
+                self.task_storage[task_id]["error"] = error_message
+                # Store HTTP error details for frontend modal
+                self.task_storage[task_id]["error_status"] = e.status_code
+                self.task_storage[task_id]["error_data"] = error_detail if isinstance(error_detail, dict) else {"error": str(error_detail)}
+                logger.info(f"[TaskManager] Stored error_status={e.status_code} and error_data keys: {list(error_detail.keys()) if isinstance(error_detail, dict) else 'not-dict'}")
+            else:
+                await self.update_progress(task_id, f"❌ Medium generation failed: {str(e)}")
+                self.task_storage[task_id]["status"] = "failed"
+                self.task_storage[task_id]["error"] = str(e)
 
 
 # Global task manager instance
diff --git a/backend/api/subscription_api.py b/backend/api/subscription_api.py
index 074a77d6..9abc7ceb 100644
--- a/backend/api/subscription_api.py
+++ b/backend/api/subscription_api.py
@@ -12,6 +12,7 @@ from functools import lru_cache
 
 from services.database import get_db
 from services.subscription import UsageTrackingService, PricingService
+from services.subscription.schema_utils import ensure_subscription_plan_columns
 from middleware.auth_middleware import get_current_user
 from models.subscription_models import (
     APIProvider, SubscriptionPlan, UserSubscription, UsageSummary,
@@ -79,6 +80,8 @@ async def get_subscription_plans(
     """Get all available subscription plans."""
     
     try:
+        # Ensure required columns exist (handles environments without migrations applied yet)
+        ensure_subscription_plan_columns(db)
         plans = db.query(SubscriptionPlan).filter(
             SubscriptionPlan.is_active == True
         ).order_by(SubscriptionPlan.price_monthly).all()
@@ -137,6 +140,7 @@ async def get_user_subscription(
         raise HTTPException(status_code=403, detail="Access denied")
     
     try:
+        ensure_subscription_plan_columns(db)
         subscription = db.query(UserSubscription).filter(
             UserSubscription.user_id == user_id,
             UserSubscription.is_active == True
@@ -234,6 +238,7 @@ async def get_subscription_status(
         raise HTTPException(status_code=403, detail="Access denied")
 
     try:
+        ensure_subscription_plan_columns(db)
         subscription = db.query(UserSubscription).filter(
             UserSubscription.user_id == user_id,
             UserSubscription.is_active == True
@@ -346,6 +351,7 @@ async def subscribe_to_plan(
         raise HTTPException(status_code=403, detail="Access denied")
 
     try:
+        ensure_subscription_plan_columns(db)
         plan_id = subscription_data.get('plan_id')
         billing_cycle = subscription_data.get('billing_cycle', 'monthly')
 
@@ -427,11 +433,16 @@ async def subscribe_to_plan(
             logger.info(f"   📊 No usage summary found for period {current_period} (will be created on reset)")
 
         # Clear subscription limits cache to force refresh on next check
+        # IMPORTANT: Do this BEFORE resetting usage to ensure cache is cleared first
         try:
             from services.subscription import PricingService
             # Clear cache for this specific user (class-level cache shared across all instances)
             cleared_count = PricingService.clear_user_cache(user_id)
             logger.info(f"   🗑️  Cleared {cleared_count} subscription cache entries for user {user_id}")
+            
+            # Also expire all SQLAlchemy objects to force fresh reads
+            db.expire_all()
+            logger.info(f"   🔄 Expired all SQLAlchemy objects to force fresh reads")
         except Exception as cache_err:
             logger.error(f"   ❌ Failed to clear cache after subscribe: {cache_err}")
 
@@ -441,12 +452,22 @@ async def subscribe_to_plan(
             usage_service = UsageTrackingService(db)
             reset_result = await usage_service.reset_current_billing_period(user_id)
             
-            # Re-query usage summary from DB after reset to get fresh data
+            # Force commit to ensure reset is persisted
+            db.commit()
+            
+            # Expire all SQLAlchemy objects to force fresh reads
+            db.expire_all()
+            
+            # Re-query usage summary from DB after reset to get fresh data (fresh query)
             usage_after = db.query(UsageSummary).filter(
                 UsageSummary.user_id == user_id,
                 UsageSummary.billing_period == current_period
             ).first()
             
+            # Refresh the usage object if found to ensure we have latest data
+            if usage_after:
+                db.refresh(usage_after)
+            
             if reset_result.get('reset'):
                 logger.info(f"   ✅ Usage counters RESET successfully")
                 if usage_after:
@@ -635,6 +656,7 @@ async def get_dashboard_data(
     """Get comprehensive dashboard data for usage monitoring."""
     
     try:
+        ensure_subscription_plan_columns(db)
         # Serve from short TTL cache to avoid hammering DB on bursts
         import time
         now = time.time()
diff --git a/backend/api/wix_routes.py b/backend/api/wix_routes.py
index 9e35f25d..5f9ac837 100644
--- a/backend/api/wix_routes.py
+++ b/backend/api/wix_routes.py
@@ -535,15 +535,33 @@ async def test_publish_real(payload: Dict[str, Any]) -> Dict[str, Any]:
         if not member_id:
             raise HTTPException(status_code=400, detail="Unable to resolve member_id from token")
 
+        # Extract SEO metadata if provided
+        seo_metadata = payload.get("seo_metadata")
+        
+        # Extract category/tag IDs or names
+        # Can be either:
+        # - IDs: List of UUID strings
+        # - Names: List of name strings (will be looked up/created)
+        category_ids = payload.get("category_ids") or payload.get("category_names")
+        tag_ids = payload.get("tag_ids") or payload.get("tag_names")
+        
+        # If SEO metadata has categories/tags but they weren't explicitly provided, use them
+        if seo_metadata:
+            if not category_ids and seo_metadata.get("blog_categories"):
+                category_ids = seo_metadata.get("blog_categories")
+            if not tag_ids and seo_metadata.get("blog_tags"):
+                tag_ids = seo_metadata.get("blog_tags")
+        
         result = wix_service.create_blog_post(
             access_token=access_token,
             title=payload.get("title") or "Untitled",
             content=payload.get("content") or "",
             cover_image_url=payload.get("cover_image_url"),
-            category_ids=payload.get("category_ids") or None,
-            tag_ids=payload.get("tag_ids") or None,
+            category_ids=category_ids,
+            tag_ids=tag_ids,
             publish=bool(payload.get("publish", True)),
             member_id=member_id,
+            seo_metadata=seo_metadata,
         )
 
         return {
diff --git a/backend/app.py b/backend/app.py
index e4a9cdca..80f673ea 100644
--- a/backend/app.py
+++ b/backend/app.py
@@ -11,6 +11,7 @@ import asyncio
 from datetime import datetime
 from services.subscription import monitoring_middleware
 
+
 # Import modular utilities
 from alwrity_utils import HealthChecker, RateLimiter, FrontendServing, RouterManager, OnboardingManager
 
diff --git a/backend/database/migrations/006_add_exa_provider.sql b/backend/database/migrations/006_add_exa_provider.sql
new file mode 100644
index 00000000..47fe3e8c
--- /dev/null
+++ b/backend/database/migrations/006_add_exa_provider.sql
@@ -0,0 +1,17 @@
+-- Add EXA to subscription plans
+ALTER TABLE subscription_plans 
+ADD COLUMN exa_calls_limit INT DEFAULT 0;
+
+-- Add EXA to usage summaries
+ALTER TABLE usage_summaries 
+ADD COLUMN exa_calls INT DEFAULT 0;
+
+ALTER TABLE usage_summaries 
+ADD COLUMN exa_cost FLOAT DEFAULT 0.0;
+
+-- Update default limits for existing plans
+UPDATE subscription_plans SET exa_calls_limit = 100 WHERE tier = 'free';
+UPDATE subscription_plans SET exa_calls_limit = 500 WHERE tier = 'basic';
+UPDATE subscription_plans SET exa_calls_limit = 2000 WHERE tier = 'pro';
+UPDATE subscription_plans SET exa_calls_limit = 0 WHERE tier = 'enterprise';
+
diff --git a/backend/models/blog_models.py b/backend/models/blog_models.py
index 51a01113..01b9ea32 100644
--- a/backend/models/blog_models.py
+++ b/backend/models/blog_models.py
@@ -1,5 +1,6 @@
 from pydantic import BaseModel, Field
 from typing import List, Optional, Dict, Any
+from enum import Enum
 
 
 class PersonaInfo(BaseModel):
@@ -50,6 +51,51 @@ class GroundingMetadata(BaseModel):
     web_search_queries: List[str] = []
 
 
+class ResearchMode(str, Enum):
+    """Research modes for different depth levels."""
+    BASIC = "basic"
+    COMPREHENSIVE = "comprehensive"
+    TARGETED = "targeted"
+
+
+class SourceType(str, Enum):
+    """Types of sources to include in research."""
+    WEB = "web"
+    ACADEMIC = "academic"
+    NEWS = "news"
+    INDUSTRY = "industry"
+    EXPERT = "expert"
+
+
+class DateRange(str, Enum):
+    """Date range filters for research."""
+    LAST_WEEK = "last_week"
+    LAST_MONTH = "last_month"
+    LAST_3_MONTHS = "last_3_months"
+    LAST_6_MONTHS = "last_6_months"
+    LAST_YEAR = "last_year"
+    ALL_TIME = "all_time"
+
+
+class ResearchProvider(str, Enum):
+    """Research provider options."""
+    GOOGLE = "google"  # Gemini native grounding
+    EXA = "exa"        # Exa neural search
+
+
+class ResearchConfig(BaseModel):
+    """Configuration for research execution."""
+    mode: ResearchMode = ResearchMode.BASIC
+    provider: ResearchProvider = ResearchProvider.GOOGLE
+    date_range: Optional[DateRange] = None
+    source_types: List[SourceType] = []
+    max_sources: int = 10
+    include_statistics: bool = True
+    include_expert_quotes: bool = True
+    include_competitors: bool = True
+    include_trends: bool = True
+
+
 class BlogResearchRequest(BaseModel):
     keywords: List[str]
     topic: Optional[str] = None
@@ -58,6 +104,8 @@ class BlogResearchRequest(BaseModel):
     tone: Optional[str] = None
     word_count_target: Optional[int] = 1500
     persona: Optional[PersonaInfo] = None
+    research_mode: Optional[ResearchMode] = ResearchMode.BASIC
+    config: Optional[ResearchConfig] = None
 
 
 class BlogResearchResponse(BaseModel):
diff --git a/backend/models/subscription_models.py b/backend/models/subscription_models.py
index 0e770ca3..a6c73945 100644
--- a/backend/models/subscription_models.py
+++ b/backend/models/subscription_models.py
@@ -34,6 +34,7 @@ class APIProvider(enum.Enum):
     METAPHOR = "metaphor"
     FIRECRAWL = "firecrawl"
     STABILITY = "stability"
+    EXA = "exa"
 
 class BillingCycle(enum.Enum):
     MONTHLY = "monthly"
@@ -66,6 +67,7 @@ class SubscriptionPlan(Base):
     metaphor_calls_limit = Column(Integer, default=0)
     firecrawl_calls_limit = Column(Integer, default=0)
     stability_calls_limit = Column(Integer, default=0)  # Image generation
+    exa_calls_limit = Column(Integer, default=0)  # Exa neural search
     
     # Token Limits (for LLM providers)
     gemini_tokens_limit = Column(Integer, default=0)
@@ -182,6 +184,7 @@ class UsageSummary(Base):
     metaphor_calls = Column(Integer, default=0)
     firecrawl_calls = Column(Integer, default=0)
     stability_calls = Column(Integer, default=0)
+    exa_calls = Column(Integer, default=0)
     
     # Token Usage
     gemini_tokens = Column(Integer, default=0)
@@ -199,6 +202,7 @@ class UsageSummary(Base):
     metaphor_cost = Column(Float, default=0.0)
     firecrawl_cost = Column(Float, default=0.0)
     stability_cost = Column(Float, default=0.0)
+    exa_cost = Column(Float, default=0.0)
     
     # Totals
     total_calls = Column(Integer, default=0)
diff --git a/backend/requirements.txt b/backend/requirements.txt
index edfa663f..76557032 100644
--- a/backend/requirements.txt
+++ b/backend/requirements.txt
@@ -38,6 +38,7 @@ aiohttp>=3.9.0
 # Data processing
 pandas>=2.0.0
 numpy>=1.24.0
+markdown>=3.5.0
 
 # SEO Analysis dependencies
 advertools>=0.14.0
diff --git a/backend/scripts/update_basic_plan_limits.py b/backend/scripts/update_basic_plan_limits.py
index e88350dc..3a5ec090 100644
--- a/backend/scripts/update_basic_plan_limits.py
+++ b/backend/scripts/update_basic_plan_limits.py
@@ -3,7 +3,7 @@ Script to update Basic plan subscription limits for testing rate limits and rene
 
 Updates:
 - LLM Calls (all providers): 10 calls (was 500-1000)
-- LLM Tokens (all providers): 2000 tokens (was 200k-1M)
+- LLM Tokens (all providers): 5000 tokens (increased from 2000 to support research workflow)
 - Images: 5 images (was 50)
 
 This script updates the SubscriptionPlan table, which automatically applies to all users
@@ -69,11 +69,11 @@ def update_basic_plan_limits():
             basic_plan.anthropic_calls_limit = 200
             basic_plan.mistral_calls_limit = 500
             
-            # Update all LLM provider token limits to 2000
-            basic_plan.gemini_tokens_limit = 2000
-            basic_plan.openai_tokens_limit = 2000
-            basic_plan.anthropic_tokens_limit = 2000
-            basic_plan.mistral_tokens_limit = 2000
+            # Update all LLM provider token limits to 20000 (increased from 5000 for better stability)
+            basic_plan.gemini_tokens_limit = 20000
+            basic_plan.openai_tokens_limit = 20000
+            basic_plan.anthropic_tokens_limit = 20000
+            basic_plan.mistral_tokens_limit = 20000
             
             # Update image generation limit to 5
             basic_plan.stability_calls_limit = 5
@@ -83,7 +83,7 @@ def update_basic_plan_limits():
             
             logger.info("\n📝 New Basic plan limits:")
             logger.info(f"  LLM Calls (all providers): 10")
-            logger.info(f"  LLM Tokens (all providers): 2000")
+            logger.info(f"  LLM Tokens (all providers): 20000 (increased from 5000)")
             logger.info(f"  Images: 5")
             
             # Count and get affected users
@@ -118,7 +118,7 @@ def update_basic_plan_limits():
             
             # New limits - use unified AI text generation limit if available
             new_call_limit = getattr(basic_plan, 'ai_text_generation_calls_limit', None) or basic_plan.gemini_calls_limit
-            new_token_limit = basic_plan.gemini_tokens_limit  # 2000
+            new_token_limit = basic_plan.gemini_tokens_limit  # 5000 (increased from 2000)
             new_image_limit = basic_plan.stability_calls_limit  # 5
             
             for sub in user_subscriptions:
@@ -253,7 +253,7 @@ if __name__ == "__main__":
     logger.info("="*60)
     logger.info("This will update Basic plan limits for testing rate limits:")
     logger.info("  - LLM Calls: 10 (all providers)")
-    logger.info("  - LLM Tokens: 2000 (all providers)")
+    logger.info("  - LLM Tokens: 20000 (all providers, increased from 5000)")
     logger.info("  - Images: 5")
     logger.info("="*60)
     
diff --git a/backend/services/blog_writer/content/medium_blog_generator.py b/backend/services/blog_writer/content/medium_blog_generator.py
index df243c4b..00231c0d 100644
--- a/backend/services/blog_writer/content/medium_blog_generator.py
+++ b/backend/services/blog_writer/content/medium_blog_generator.py
@@ -8,6 +8,7 @@ import time
 import json
 from typing import Dict, Any, List
 from loguru import logger
+from fastapi import HTTPException
 
 from models.blog_models import (
     MediumBlogGenerateRequest,
@@ -25,8 +26,20 @@ class MediumBlogGenerator:
     def __init__(self):
         self.cache = persistent_content_cache
     
-    async def generate_medium_blog_with_progress(self, req: MediumBlogGenerateRequest, task_id: str) -> MediumBlogGenerateResult:
-        """Use Gemini structured JSON to generate a medium-length blog in one call."""
+    async def generate_medium_blog_with_progress(self, req: MediumBlogGenerateRequest, task_id: str, user_id: str) -> MediumBlogGenerateResult:
+        """Use Gemini structured JSON to generate a medium-length blog in one call.
+        
+        Args:
+            req: Medium blog generation request
+            task_id: Task ID for progress updates
+            user_id: User ID (required for subscription checks and usage tracking)
+            
+        Raises:
+            ValueError: If user_id is not provided
+        """
+        if not user_id:
+            raise ValueError("user_id is required for medium blog generation (subscription checks and usage tracking)")
+        
         import time
         start = time.time()
 
@@ -156,7 +169,7 @@ class MediumBlogGenerator:
             - Use language that resonates with {audience}
             - Maintain consistent voice that reflects this persona's expertise
             """
-        
+
         prompt = (
             f"Write blog content for the following sections. Each section should be {req.globalTargetWords or 1000} words total, distributed across all sections.\n\n"
             f"Blog Title: {req.title}\n\n"
@@ -176,11 +189,20 @@ class MediumBlogGenerator:
             f"Sections to write:\n{json.dumps(payload, ensure_ascii=False, indent=2)}"
         )
 
-        ai_resp = llm_text_gen(
-            prompt=prompt,
-            json_struct=schema,
-            system_prompt=system,
-        )
+        try:
+            ai_resp = llm_text_gen(
+                prompt=prompt,
+                json_struct=schema,
+                system_prompt=system,
+                user_id=user_id
+            )
+        except HTTPException:
+            # Re-raise HTTPExceptions (e.g., 429 subscription limit) to preserve error details
+            raise
+        except Exception as llm_error:
+            # Wrap other errors
+            logger.error(f"AI generation failed: {llm_error}")
+            raise Exception(f"AI generation failed: {str(llm_error)}")
 
         # Check for errors in AI response
         if not ai_resp or ai_resp.get("error"):
diff --git a/backend/services/blog_writer/core/blog_writer_service.py b/backend/services/blog_writer/core/blog_writer_service.py
index b8324420..89b7c919 100644
--- a/backend/services/blog_writer/core/blog_writer_service.py
+++ b/backend/services/blog_writer/core/blog_writer_service.py
@@ -105,13 +105,20 @@ class BlogWriterService:
         return await self.research_service.research_with_progress(request, task_id, user_id)
     
     # Outline Methods
-    async def generate_outline(self, request: BlogOutlineRequest) -> BlogOutlineResponse:
-        """Generate AI-powered outline from research data."""
-        return await self.outline_service.generate_outline(request)
+    async def generate_outline(self, request: BlogOutlineRequest, user_id: str) -> BlogOutlineResponse:
+        """Generate AI-powered outline from research data.
+        
+        Args:
+            request: Outline generation request with research data
+            user_id: User ID (required for subscription checks and usage tracking)
+        """
+        if not user_id:
+            raise ValueError("user_id is required for outline generation (subscription checks and usage tracking)")
+        return await self.outline_service.generate_outline(request, user_id)
     
-    async def generate_outline_with_progress(self, request: BlogOutlineRequest, task_id: str) -> BlogOutlineResponse:
+    async def generate_outline_with_progress(self, request: BlogOutlineRequest, task_id: str, user_id: str) -> BlogOutlineResponse:
         """Generate outline with real-time progress updates."""
-        return await self.outline_service.generate_outline_with_progress(request, task_id)
+        return await self.outline_service.generate_outline_with_progress(request, task_id, user_id)
     
     async def refine_outline(self, request: BlogOutlineRefineRequest) -> BlogOutlineResponse:
         """Refine outline with HITL operations."""
@@ -334,9 +341,17 @@ class BlogWriterService:
         # TODO: Move to content module
         return BlogPublishResponse(success=True, platform=request.platform, url="https://example.com/post")
 
-    async def generate_medium_blog_with_progress(self, req: MediumBlogGenerateRequest, task_id: str) -> MediumBlogGenerateResult:
-        """Use Gemini structured JSON to generate a medium-length blog in one call."""
-        return await self.medium_blog_generator.generate_medium_blog_with_progress(req, task_id)
+    async def generate_medium_blog_with_progress(self, req: MediumBlogGenerateRequest, task_id: str, user_id: str) -> MediumBlogGenerateResult:
+        """Use Gemini structured JSON to generate a medium-length blog in one call.
+        
+        Args:
+            req: Medium blog generation request
+            task_id: Task ID for progress updates
+            user_id: User ID (required for subscription checks and usage tracking)
+        """
+        if not user_id:
+            raise ValueError("user_id is required for medium blog generation (subscription checks and usage tracking)")
+        return await self.medium_blog_generator.generate_medium_blog_with_progress(req, task_id, user_id)
 
     async def analyze_flow_basic(self, request: Dict[str, Any]) -> Dict[str, Any]:
         """Analyze flow metrics for entire blog using single AI call (cost-effective)."""
diff --git a/backend/services/blog_writer/outline/outline_generator.py b/backend/services/blog_writer/outline/outline_generator.py
index a99ed5c3..40bfe0ae 100644
--- a/backend/services/blog_writer/outline/outline_generator.py
+++ b/backend/services/blog_writer/outline/outline_generator.py
@@ -42,10 +42,20 @@ class OutlineGenerator:
         self.response_processor = ResponseProcessor()
         self.parallel_processor = ParallelProcessor(self.source_mapper, self.grounding_engine)
     
-    async def generate(self, request: BlogOutlineRequest) -> BlogOutlineResponse:
+    async def generate(self, request: BlogOutlineRequest, user_id: str) -> BlogOutlineResponse:
         """
-        Generate AI-powered outline using research results
+        Generate AI-powered outline using research results.
+        
+        Args:
+            request: Outline generation request with research data
+            user_id: User ID (required for subscription checks and usage tracking)
+            
+        Raises:
+            ValueError: If user_id is not provided
         """
+        if not user_id:
+            raise ValueError("user_id is required for outline generation (subscription checks and usage tracking)")
+        
         # Extract research insights
         research = request.research
         primary_keywords = research.keyword_analysis.get('primary', [])
@@ -68,15 +78,15 @@ class OutlineGenerator:
         # Define schema with proper property ordering (critical for Gemini API)
         outline_schema = self.prompt_builder.get_outline_schema()
         
-        # Generate outline using structured JSON response with retry logic
-        outline_data = await self.response_processor.generate_with_retry(outline_prompt, outline_schema)
+        # Generate outline using structured JSON response with retry logic (user_id required)
+        outline_data = await self.response_processor.generate_with_retry(outline_prompt, outline_schema, user_id)
         
         # Convert to BlogOutlineSection objects
         outline_sections = self.response_processor.convert_to_sections(outline_data, sources)
         
-        # Run parallel processing for speed optimization
+        # Run parallel processing for speed optimization (user_id required)
         mapped_sections, grounding_insights = await self.parallel_processor.run_parallel_processing_async(
-            outline_sections, research
+            outline_sections, research, user_id
         )
         
         # Enhance sections with grounding insights
@@ -85,9 +95,9 @@ class OutlineGenerator:
             mapped_sections, research.grounding_metadata, grounding_insights
         )
         
-        # Optimize outline for better flow, SEO, and engagement
+        # Optimize outline for better flow, SEO, and engagement (user_id required)
         logger.info("Optimizing outline for better flow and engagement...")
-        optimized_sections = await self.outline_optimizer.optimize(grounding_enhanced_sections, "comprehensive optimization")
+        optimized_sections = await self.outline_optimizer.optimize(grounding_enhanced_sections, "comprehensive optimization", user_id)
         
         # Rebalance word counts for optimal distribution
         target_words = request.word_count or 1500
@@ -118,10 +128,21 @@ class OutlineGenerator:
             research_coverage=research_coverage
         )
     
-    async def generate_with_progress(self, request: BlogOutlineRequest, task_id: str) -> BlogOutlineResponse:
+    async def generate_with_progress(self, request: BlogOutlineRequest, task_id: str, user_id: str) -> BlogOutlineResponse:
         """
         Outline generation method with progress updates for real-time feedback.
+        
+        Args:
+            request: Outline generation request with research data
+            task_id: Task ID for progress updates
+            user_id: User ID (required for subscription checks and usage tracking)
+            
+        Raises:
+            ValueError: If user_id is not provided
         """
+        if not user_id:
+            raise ValueError("user_id is required for outline generation (subscription checks and usage tracking)")
+        
         from api.blog_writer.task_manager import task_manager
         
         # Extract research insights
@@ -150,17 +171,17 @@ class OutlineGenerator:
         
         await task_manager.update_progress(task_id, "🔄 Making AI request to generate structured outline...")
         
-        # Generate outline using structured JSON response with retry logic
-        outline_data = await self.response_processor.generate_with_retry(outline_prompt, outline_schema, task_id)
+        # Generate outline using structured JSON response with retry logic (user_id required for subscription checks)
+        outline_data = await self.response_processor.generate_with_retry(outline_prompt, outline_schema, user_id, task_id)
         
         await task_manager.update_progress(task_id, "📝 Processing outline structure and validating sections...")
         
         # Convert to BlogOutlineSection objects
         outline_sections = self.response_processor.convert_to_sections(outline_data, sources)
         
-        # Run parallel processing for speed optimization
+        # Run parallel processing for speed optimization (user_id required for subscription checks)
         mapped_sections, grounding_insights = await self.parallel_processor.run_parallel_processing(
-            outline_sections, research, task_id
+            outline_sections, research, user_id, task_id
         )
         
         # Enhance sections with grounding insights (depends on both previous tasks)
@@ -169,9 +190,9 @@ class OutlineGenerator:
             mapped_sections, research.grounding_metadata, grounding_insights
         )
         
-        # Optimize outline for better flow, SEO, and engagement
+        # Optimize outline for better flow, SEO, and engagement (user_id required for subscription checks)
         await task_manager.update_progress(task_id, "🎯 Optimizing outline for better flow and engagement...")
-        optimized_sections = await self.outline_optimizer.optimize(grounding_enhanced_sections, "comprehensive optimization")
+        optimized_sections = await self.outline_optimizer.optimize(grounding_enhanced_sections, "comprehensive optimization", user_id)
         
         # Rebalance word counts for optimal distribution
         await task_manager.update_progress(task_id, "⚖️ Rebalancing word count distribution...")
diff --git a/backend/services/blog_writer/outline/outline_optimizer.py b/backend/services/blog_writer/outline/outline_optimizer.py
index e5c4c0fb..8fa36e2a 100644
--- a/backend/services/blog_writer/outline/outline_optimizer.py
+++ b/backend/services/blog_writer/outline/outline_optimizer.py
@@ -13,8 +13,23 @@ from models.blog_models import BlogOutlineSection
 class OutlineOptimizer:
     """Optimizes outlines for better flow, SEO, and engagement."""
     
-    async def optimize(self, outline: List[BlogOutlineSection], focus: str = "general optimization") -> List[BlogOutlineSection]:
-        """Optimize entire outline for better flow, SEO, and engagement."""
+    async def optimize(self, outline: List[BlogOutlineSection], focus: str, user_id: str) -> List[BlogOutlineSection]:
+        """Optimize entire outline for better flow, SEO, and engagement.
+        
+        Args:
+            outline: List of outline sections to optimize
+            focus: Optimization focus (e.g., "general optimization")
+            user_id: User ID (required for subscription checks and usage tracking)
+            
+        Returns:
+            List of optimized outline sections
+            
+        Raises:
+            ValueError: If user_id is not provided
+        """
+        if not user_id:
+            raise ValueError("user_id is required for outline optimization (subscription checks and usage tracking)")
+        
         outline_text = "\n".join([f"{i+1}. {s.heading}" for i, s in enumerate(outline)])
         
         optimization_prompt = f"""Optimize this blog outline for better flow, engagement, and SEO:
@@ -67,7 +82,8 @@ Return JSON format:
             optimized_data = llm_text_gen(
                 prompt=optimization_prompt,
                 json_struct=optimization_schema,
-                system_prompt=None
+                system_prompt=None,
+                user_id=user_id
             )
             
             # Handle the new schema format with "outline" wrapper
diff --git a/backend/services/blog_writer/outline/outline_service.py b/backend/services/blog_writer/outline/outline_service.py
index 2dc6d506..1118b3de 100644
--- a/backend/services/blog_writer/outline/outline_service.py
+++ b/backend/services/blog_writer/outline/outline_service.py
@@ -29,11 +29,21 @@ class OutlineService:
         self.outline_optimizer = OutlineOptimizer()
         self.section_enhancer = SectionEnhancer()
     
-    async def generate_outline(self, request: BlogOutlineRequest) -> BlogOutlineResponse:
+    async def generate_outline(self, request: BlogOutlineRequest, user_id: str) -> BlogOutlineResponse:
         """
-        Stage 2: Content Planning with AI-generated outline using research results
-        Uses Gemini with research data to create comprehensive, SEO-optimized outline
+        Stage 2: Content Planning with AI-generated outline using research results.
+        Uses Gemini with research data to create comprehensive, SEO-optimized outline.
+        
+        Args:
+            request: Outline generation request with research data
+            user_id: User ID (required for subscription checks and usage tracking)
+            
+        Raises:
+            ValueError: If user_id is not provided
         """
+        if not user_id:
+            raise ValueError("user_id is required for outline generation (subscription checks and usage tracking)")
+        
         # Extract cache parameters - use original user keywords for consistent caching
         keywords = request.research.original_keywords or request.research.keyword_analysis.get('primary', [])
         industry = getattr(request.persona, 'industry', 'general') if request.persona else 'general'
@@ -56,9 +66,9 @@ class OutlineService:
             logger.info(f"Using cached outline for keywords: {keywords}")
             return BlogOutlineResponse(**cached_result)
         
-        # Generate new outline if not cached
+        # Generate new outline if not cached (user_id required)
         logger.info(f"Generating new outline for keywords: {keywords}")
-        result = await self.outline_generator.generate(request)
+        result = await self.outline_generator.generate(request, user_id)
         
         # Cache the result
         persistent_outline_cache.cache_outline(
@@ -73,7 +83,7 @@ class OutlineService:
         
         return result
     
-    async def generate_outline_with_progress(self, request: BlogOutlineRequest, task_id: str) -> BlogOutlineResponse:
+    async def generate_outline_with_progress(self, request: BlogOutlineRequest, task_id: str, user_id: str) -> BlogOutlineResponse:
         """
         Outline generation method with progress updates for real-time feedback.
         """
@@ -104,7 +114,7 @@ class OutlineService:
         
         # Generate new outline if not cached
         logger.info(f"Generating new outline for keywords: {keywords} (with progress updates)")
-        result = await self.outline_generator.generate_with_progress(request, task_id)
+        result = await self.outline_generator.generate_with_progress(request, task_id, user_id)
         
         # Cache the result
         persistent_outline_cache.cache_outline(
diff --git a/backend/services/blog_writer/outline/parallel_processor.py b/backend/services/blog_writer/outline/parallel_processor.py
index 395f8dda..61e066ed 100644
--- a/backend/services/blog_writer/outline/parallel_processor.py
+++ b/backend/services/blog_writer/outline/parallel_processor.py
@@ -17,18 +17,25 @@ class ParallelProcessor:
         self.source_mapper = source_mapper
         self.grounding_engine = grounding_engine
     
-    async def run_parallel_processing(self, outline_sections, research, task_id: str = None) -> Tuple[Any, Any]:
+    async def run_parallel_processing(self, outline_sections, research, user_id: str, task_id: str = None) -> Tuple[Any, Any]:
         """
         Run source mapping and grounding insights extraction in parallel.
         
         Args:
             outline_sections: List of outline sections to process
             research: Research data object
+            user_id: User ID (required for subscription checks and usage tracking)
             task_id: Optional task ID for progress updates
             
         Returns:
             Tuple of (mapped_sections, grounding_insights)
+            
+        Raises:
+            ValueError: If user_id is not provided
         """
+        if not user_id:
+            raise ValueError("user_id is required for parallel processing (subscription checks and usage tracking)")
+        
         if task_id:
             from api.blog_writer.task_manager import task_manager
             await task_manager.update_progress(task_id, "⚡ Running parallel processing for maximum speed...")
@@ -37,7 +44,7 @@ class ParallelProcessor:
         
         # Run these tasks in parallel to save time
         source_mapping_task = asyncio.create_task(
-            self._run_source_mapping(outline_sections, research, task_id)
+            self._run_source_mapping(outline_sections, research, task_id, user_id)
         )
         
         grounding_insights_task = asyncio.create_task(
@@ -52,22 +59,29 @@ class ParallelProcessor:
         
         return mapped_sections, grounding_insights
     
-    async def run_parallel_processing_async(self, outline_sections, research) -> Tuple[Any, Any]:
+    async def run_parallel_processing_async(self, outline_sections, research, user_id: str) -> Tuple[Any, Any]:
         """
         Run parallel processing without progress updates (for non-progress methods).
         
         Args:
             outline_sections: List of outline sections to process
             research: Research data object
+            user_id: User ID (required for subscription checks and usage tracking)
             
         Returns:
             Tuple of (mapped_sections, grounding_insights)
+            
+        Raises:
+            ValueError: If user_id is not provided
         """
+        if not user_id:
+            raise ValueError("user_id is required for parallel processing (subscription checks and usage tracking)")
+        
         logger.info("Running parallel processing for maximum speed...")
         
         # Run these tasks in parallel to save time
         source_mapping_task = asyncio.create_task(
-            self._run_source_mapping_async(outline_sections, research)
+            self._run_source_mapping_async(outline_sections, research, user_id)
         )
         
         grounding_insights_task = asyncio.create_task(
@@ -82,12 +96,12 @@ class ParallelProcessor:
         
         return mapped_sections, grounding_insights
     
-    async def _run_source_mapping(self, outline_sections, research, task_id):
+    async def _run_source_mapping(self, outline_sections, research, task_id, user_id: str):
         """Run source mapping in parallel."""
         if task_id:
             from api.blog_writer.task_manager import task_manager
             await task_manager.update_progress(task_id, "🔗 Applying intelligent source-to-section mapping...")
-        return self.source_mapper.map_sources_to_sections(outline_sections, research)
+        return self.source_mapper.map_sources_to_sections(outline_sections, research, user_id)
     
     async def _run_grounding_insights_extraction(self, research, task_id):
         """Run grounding insights extraction in parallel."""
@@ -96,10 +110,10 @@ class ParallelProcessor:
             await task_manager.update_progress(task_id, "🧠 Extracting grounding metadata insights...")
         return self.grounding_engine.extract_contextual_insights(research.grounding_metadata)
     
-    async def _run_source_mapping_async(self, outline_sections, research):
+    async def _run_source_mapping_async(self, outline_sections, research, user_id: str):
         """Run source mapping in parallel (async version without progress updates)."""
         logger.info("Applying intelligent source-to-section mapping...")
-        return self.source_mapper.map_sources_to_sections(outline_sections, research)
+        return self.source_mapper.map_sources_to_sections(outline_sections, research, user_id)
     
     async def _run_grounding_insights_extraction_async(self, research):
         """Run grounding insights extraction in parallel (async version without progress updates)."""
diff --git a/backend/services/blog_writer/outline/response_processor.py b/backend/services/blog_writer/outline/response_processor.py
index c8d8479c..826d3509 100644
--- a/backend/services/blog_writer/outline/response_processor.py
+++ b/backend/services/blog_writer/outline/response_processor.py
@@ -18,8 +18,21 @@ class ResponseProcessor:
         """Initialize the response processor."""
         pass
     
-    async def generate_with_retry(self, prompt: str, schema: Dict[str, Any], task_id: str = None) -> Dict[str, Any]:
-        """Generate outline with retry logic for API failures."""
+    async def generate_with_retry(self, prompt: str, schema: Dict[str, Any], user_id: str, task_id: str = None) -> Dict[str, Any]:
+        """Generate outline with retry logic for API failures.
+        
+        Args:
+            prompt: The prompt for outline generation
+            schema: JSON schema for structured response
+            user_id: User ID (required for subscription checks and usage tracking)
+            task_id: Optional task ID for progress updates
+            
+        Raises:
+            ValueError: If user_id is not provided
+        """
+        if not user_id:
+            raise ValueError("user_id is required for outline generation (subscription checks and usage tracking)")
+        
         from services.llm_providers.main_text_generation import llm_text_gen
         from api.blog_writer.task_manager import task_manager
         
@@ -34,7 +47,8 @@ class ResponseProcessor:
                 outline_data = llm_text_gen(
                     prompt=prompt,
                     json_struct=schema,
-                    system_prompt=None
+                    system_prompt=None,
+                    user_id=user_id
                 )
                 
                 # Log response for debugging
diff --git a/backend/services/blog_writer/outline/section_enhancer.py b/backend/services/blog_writer/outline/section_enhancer.py
index 936576bc..8cd47890 100644
--- a/backend/services/blog_writer/outline/section_enhancer.py
+++ b/backend/services/blog_writer/outline/section_enhancer.py
@@ -12,8 +12,23 @@ from models.blog_models import BlogOutlineSection
 class SectionEnhancer:
     """Enhances individual outline sections using AI."""
     
-    async def enhance(self, section: BlogOutlineSection, focus: str = "general improvement") -> BlogOutlineSection:
-        """Enhance a section using AI with research context."""
+    async def enhance(self, section: BlogOutlineSection, focus: str, user_id: str) -> BlogOutlineSection:
+        """Enhance a section using AI with research context.
+        
+        Args:
+            section: Outline section to enhance
+            focus: Enhancement focus (e.g., "general improvement")
+            user_id: User ID (required for subscription checks and usage tracking)
+            
+        Returns:
+            Enhanced outline section
+            
+        Raises:
+            ValueError: If user_id is not provided
+        """
+        if not user_id:
+            raise ValueError("user_id is required for section enhancement (subscription checks and usage tracking)")
+        
         enhancement_prompt = f"""
         Enhance the following blog section to make it more engaging, comprehensive, and valuable:
         
@@ -61,7 +76,8 @@ class SectionEnhancer:
             enhanced_data = llm_text_gen(
                 prompt=enhancement_prompt,
                 json_struct=enhancement_schema,
-                system_prompt=None
+                system_prompt=None,
+                user_id=user_id
             )
             
             if isinstance(enhanced_data, dict) and 'error' not in enhanced_data:
diff --git a/backend/services/blog_writer/outline/source_mapper.py b/backend/services/blog_writer/outline/source_mapper.py
index d630b22b..8ff766be 100644
--- a/backend/services/blog_writer/outline/source_mapper.py
+++ b/backend/services/blog_writer/outline/source_mapper.py
@@ -52,7 +52,8 @@ class SourceToSectionMapper:
     def map_sources_to_sections(
         self, 
         sections: List[BlogOutlineSection], 
-        research_data: BlogResearchResponse
+        research_data: BlogResearchResponse,
+        user_id: str
     ) -> List[BlogOutlineSection]:
         """
         Map research sources to outline sections using intelligent algorithms.
@@ -60,10 +61,17 @@ class SourceToSectionMapper:
         Args:
             sections: List of outline sections to map sources to
             research_data: Research data containing sources and metadata
+            user_id: User ID (required for subscription checks and usage tracking)
             
         Returns:
             List of outline sections with intelligently mapped sources
+            
+        Raises:
+            ValueError: If user_id is not provided
         """
+        if not user_id:
+            raise ValueError("user_id is required for source mapping (subscription checks and usage tracking)")
+        
         if not sections or not research_data.sources:
             logger.warning("No sections or sources to map")
             return sections
@@ -73,8 +81,8 @@ class SourceToSectionMapper:
         # Step 1: Algorithmic mapping
         mapping_results = self._algorithmic_source_mapping(sections, research_data)
         
-        # Step 2: AI validation and improvement (single prompt)
-        validated_mapping = self._ai_validate_mapping(mapping_results, research_data)
+        # Step 2: AI validation and improvement (single prompt, user_id required for subscription checks)
+        validated_mapping = self._ai_validate_mapping(mapping_results, research_data, user_id)
         
         # Step 3: Apply validated mapping to sections
         mapped_sections = self._apply_mapping_to_sections(sections, validated_mapping)
@@ -261,7 +269,8 @@ class SourceToSectionMapper:
     def _ai_validate_mapping(
         self, 
         mapping_results: Dict[str, List[Tuple[ResearchSource, float]]], 
-        research_data: BlogResearchResponse
+        research_data: BlogResearchResponse,
+        user_id: str
     ) -> Dict[str, List[Tuple[ResearchSource, float]]]:
         """
         Use AI to validate and improve the algorithmic mapping results.
@@ -269,18 +278,25 @@ class SourceToSectionMapper:
         Args:
             mapping_results: Algorithmic mapping results
             research_data: Research data for context
+            user_id: User ID (required for subscription checks and usage tracking)
             
         Returns:
             AI-validated and improved mapping results
+            
+        Raises:
+            ValueError: If user_id is not provided
         """
+        if not user_id:
+            raise ValueError("user_id is required for AI validation (subscription checks and usage tracking)")
+        
         try:
             logger.info("Starting AI validation of source-to-section mapping...")
             
             # Build AI validation prompt
             validation_prompt = self._build_validation_prompt(mapping_results, research_data)
             
-            # Get AI validation response
-            validation_response = self._get_ai_validation_response(validation_prompt)
+            # Get AI validation response (user_id required for subscription checks)
+            validation_response = self._get_ai_validation_response(validation_prompt, user_id)
             
             # Parse and apply AI validation results
             validated_mapping = self._parse_validation_response(validation_response, mapping_results, research_data)
@@ -548,23 +564,31 @@ Analyze the mapping and provide your recommendations.
         
         return prompt
     
-    def _get_ai_validation_response(self, prompt: str) -> str:
+    def _get_ai_validation_response(self, prompt: str, user_id: str) -> str:
         """
         Get AI validation response using LLM provider.
         
         Args:
             prompt: Validation prompt
+            user_id: User ID (required for subscription checks and usage tracking)
             
         Returns:
             AI validation response
+            
+        Raises:
+            ValueError: If user_id is not provided
         """
+        if not user_id:
+            raise ValueError("user_id is required for AI validation response (subscription checks and usage tracking)")
+        
         try:
             from services.llm_providers.main_text_generation import llm_text_gen
             
             response = llm_text_gen(
                 prompt=prompt,
                 json_struct=None,
-                system_prompt=None
+                system_prompt=None,
+                user_id=user_id
             )
             
             return response
diff --git a/backend/services/blog_writer/research/__init__.py b/backend/services/blog_writer/research/__init__.py
index ecf76b11..87035e12 100644
--- a/backend/services/blog_writer/research/__init__.py
+++ b/backend/services/blog_writer/research/__init__.py
@@ -13,11 +13,17 @@ from .keyword_analyzer import KeywordAnalyzer
 from .competitor_analyzer import CompetitorAnalyzer
 from .content_angle_generator import ContentAngleGenerator
 from .data_filter import ResearchDataFilter
+from .base_provider import ResearchProvider as BaseResearchProvider
+from .google_provider import GoogleResearchProvider
+from .exa_provider import ExaResearchProvider
 
 __all__ = [
     'ResearchService',
     'KeywordAnalyzer', 
     'CompetitorAnalyzer',
     'ContentAngleGenerator',
-    'ResearchDataFilter'
+    'ResearchDataFilter',
+    'BaseResearchProvider',
+    'GoogleResearchProvider',
+    'ExaResearchProvider',
 ]
diff --git a/backend/services/blog_writer/research/base_provider.py b/backend/services/blog_writer/research/base_provider.py
new file mode 100644
index 00000000..72aae9b2
--- /dev/null
+++ b/backend/services/blog_writer/research/base_provider.py
@@ -0,0 +1,37 @@
+"""
+Base Research Provider Interface
+
+Abstract base class for research provider implementations.
+Ensures consistency across different research providers (Google, Exa, etc.)
+"""
+
+from abc import ABC, abstractmethod
+from typing import Dict, Any
+
+
+class ResearchProvider(ABC):
+    """Abstract base class for research providers."""
+    
+    @abstractmethod
+    async def search(
+        self,
+        prompt: str,
+        topic: str,
+        industry: str,
+        target_audience: str,
+        config: Any,  # ResearchConfig
+        user_id: str
+    ) -> Dict[str, Any]:
+        """Execute research and return raw results."""
+        pass
+    
+    @abstractmethod
+    def get_provider_enum(self):
+        """Return APIProvider enum for subscription tracking."""
+        pass
+    
+    @abstractmethod
+    def estimate_tokens(self) -> int:
+        """Estimate token usage for pre-flight validation."""
+        pass
+
diff --git a/backend/services/blog_writer/research/exa_provider.py b/backend/services/blog_writer/research/exa_provider.py
new file mode 100644
index 00000000..6c20a0d6
--- /dev/null
+++ b/backend/services/blog_writer/research/exa_provider.py
@@ -0,0 +1,188 @@
+"""
+Exa Research Provider
+
+Neural search implementation using Exa API for high-quality, citation-rich research.
+"""
+
+from exa_py import Exa
+import os
+from loguru import logger
+from models.subscription_models import APIProvider
+from .base_provider import ResearchProvider as BaseProvider
+
+
+class ExaResearchProvider(BaseProvider):
+    """Exa neural search provider."""
+    
+    def __init__(self):
+        self.api_key = os.getenv("EXA_API_KEY")
+        if not self.api_key:
+            raise RuntimeError("EXA_API_KEY not configured")
+        self.exa = Exa(self.api_key)
+        logger.info("✅ Exa Research Provider initialized")
+    
+    async def search(self, prompt, topic, industry, target_audience, config, user_id):
+        """Execute Exa neural search and return standardized results."""
+        # Build Exa query
+        query = f"{topic} {industry} {target_audience}"
+        
+        # Map source types to Exa categories
+        category = self._map_source_type_to_category(config.source_types)
+        
+        logger.info(f"[Exa Research] Executing search: {query}")
+        
+        # Execute Exa search
+        results = self.exa.search_and_contents(
+            query,
+            type="auto",
+            category=category,
+            num_results=min(config.max_sources, 25),
+            contents={
+                'text': {'max_characters': 1000},
+                'summary': {'query': f"Key insights about {topic}"},
+                'highlights': {
+                    'num_sentences': 2,
+                    'highlights_per_url': 3
+                }
+            }
+        )
+        
+        # Transform to standardized format
+        sources = self._transform_sources(results.results)
+        content = self._aggregate_content(results.results)
+        search_type = getattr(results, 'resolvedSearchType', 'neural') if hasattr(results, 'resolvedSearchType') else 'neural'
+        
+        # Get cost if available
+        cost = 0.005  # Default Exa cost for 1-25 results
+        if hasattr(results, 'costDollars'):
+            if hasattr(results.costDollars, 'total'):
+                cost = results.costDollars.total
+        
+        logger.info(f"[Exa Research] Search completed: {len(sources)} sources, type: {search_type}")
+        
+        return {
+            'sources': sources,
+            'content': content,
+            'search_type': search_type,
+            'provider': 'exa',
+            'search_queries': [query],
+            'cost': {'total': cost}
+        }
+    
+    def get_provider_enum(self):
+        """Return EXA provider enum for subscription tracking."""
+        return APIProvider.EXA
+    
+    def estimate_tokens(self) -> int:
+        """Estimate token usage for Exa (not token-based)."""
+        return 0  # Exa is per-search, not token-based
+    
+    def _map_source_type_to_category(self, source_types):
+        """Map SourceType enum to Exa category parameter."""
+        if not source_types:
+            return None
+        
+        category_map = {
+            'research paper': 'research paper',
+            'news': 'news',
+            'web': 'personal site',
+            'industry': 'company',
+            'expert': 'linkedin profile'
+        }
+        
+        for st in source_types:
+            if st.value in category_map:
+                return category_map[st.value]
+        
+        return None
+    
+    def _transform_sources(self, results):
+        """Transform Exa results to ResearchSource format."""
+        sources = []
+        for idx, result in enumerate(results):
+            source_type = self._determine_source_type(result.url if hasattr(result, 'url') else '')
+            
+            sources.append({
+                'title': result.title if hasattr(result, 'title') else '',
+                'url': result.url if hasattr(result, 'url') else '',
+                'excerpt': self._get_excerpt(result),
+                'credibility_score': 0.85,  # Exa results are high quality
+                'published_at': result.publishedDate if hasattr(result, 'publishedDate') else None,
+                'index': idx,
+                'source_type': source_type,
+                'content': result.text if hasattr(result, 'text') else '',
+                'highlights': result.highlights if hasattr(result, 'highlights') else [],
+                'summary': result.summary if hasattr(result, 'summary') else ''
+            })
+        
+        return sources
+    
+    def _get_excerpt(self, result):
+        """Extract excerpt from Exa result."""
+        if hasattr(result, 'text') and result.text:
+            return result.text[:500]
+        elif hasattr(result, 'summary') and result.summary:
+            return result.summary
+        return ''
+    
+    def _determine_source_type(self, url):
+        """Determine source type from URL."""
+        if not url:
+            return 'web'
+        
+        url_lower = url.lower()
+        if 'arxiv.org' in url_lower or 'research' in url_lower:
+            return 'academic'
+        elif any(news in url_lower for news in ['cnn.com', 'bbc.com', 'reuters.com', 'theguardian.com']):
+            return 'news'
+        elif 'linkedin.com' in url_lower:
+            return 'expert'
+        else:
+            return 'web'
+    
+    def _aggregate_content(self, results):
+        """Aggregate content from Exa results for LLM analysis."""
+        content_parts = []
+        
+        for idx, result in enumerate(results):
+            if hasattr(result, 'summary') and result.summary:
+                content_parts.append(f"Source {idx + 1}: {result.summary}")
+            elif hasattr(result, 'text') and result.text:
+                content_parts.append(f"Source {idx + 1}: {result.text[:1000]}")
+        
+        return "\n\n".join(content_parts)
+    
+    def track_exa_usage(self, user_id: str, cost: float):
+        """Track Exa API usage after successful call."""
+        from services.database import get_db
+        from services.subscription import PricingService
+        from sqlalchemy import text
+        
+        db = next(get_db())
+        try:
+            pricing_service = PricingService(db)
+            current_period = pricing_service.get_current_billing_period(user_id)
+            
+            # Update exa_calls and exa_cost via SQL UPDATE
+            update_query = text("""
+                UPDATE usage_summaries 
+                SET exa_calls = COALESCE(exa_calls, 0) + 1,
+                    exa_cost = COALESCE(exa_cost, 0) + :cost,
+                    total_calls = total_calls + 1,
+                    total_cost = total_cost + :cost
+                WHERE user_id = :user_id AND billing_period = :period
+            """)
+            db.execute(update_query, {
+                'cost': cost,
+                'user_id': user_id,
+                'period': current_period
+            })
+            db.commit()
+            
+            logger.info(f"[Exa] Tracked usage: user={user_id}, cost=${cost}")
+        except Exception as e:
+            logger.error(f"[Exa] Failed to track usage: {e}")
+            db.rollback()
+        finally:
+            db.close()
+
diff --git a/backend/services/blog_writer/research/google_provider.py b/backend/services/blog_writer/research/google_provider.py
new file mode 100644
index 00000000..b0aa06d8
--- /dev/null
+++ b/backend/services/blog_writer/research/google_provider.py
@@ -0,0 +1,40 @@
+"""
+Google Research Provider
+
+Wrapper for Gemini native Google Search grounding to match base provider interface.
+"""
+
+from services.llm_providers.gemini_grounded_provider import GeminiGroundedProvider
+from models.subscription_models import APIProvider
+from .base_provider import ResearchProvider as BaseProvider
+from loguru import logger
+
+
+class GoogleResearchProvider(BaseProvider):
+    """Google research provider using Gemini native grounding."""
+    
+    def __init__(self):
+        self.gemini = GeminiGroundedProvider()
+    
+    async def search(self, prompt, topic, industry, target_audience, config, user_id):
+        """Call Gemini grounding with pre-flight validation."""
+        logger.info(f"[Google Research] Executing search for topic: {topic}")
+        
+        result = await self.gemini.generate_grounded_content(
+            prompt=prompt,
+            content_type="research",
+            max_tokens=2000,
+            user_id=user_id,
+            validate_subsequent_operations=True
+        )
+        
+        return result
+    
+    def get_provider_enum(self):
+        """Return GEMINI provider enum for subscription tracking."""
+        return APIProvider.GEMINI
+    
+    def estimate_tokens(self) -> int:
+        """Estimate token usage for Google grounding."""
+        return 1200  # Conservative estimate
+
diff --git a/backend/services/blog_writer/research/research_service.py b/backend/services/blog_writer/research/research_service.py
index f7c0e2d4..533fab13 100644
--- a/backend/services/blog_writer/research/research_service.py
+++ b/backend/services/blog_writer/research/research_service.py
@@ -16,6 +16,9 @@ from models.blog_models import (
     GroundingChunk,
     GroundingSupport,
     Citation,
+    ResearchConfig,
+    ResearchMode,
+    ResearchProvider,
 )
 from services.blog_writer.logger_config import blog_writer_logger, log_function_call
 from fastapi import HTTPException
@@ -24,6 +27,7 @@ from .keyword_analyzer import KeywordAnalyzer
 from .competitor_analyzer import CompetitorAnalyzer
 from .content_angle_generator import ContentAngleGenerator
 from .data_filter import ResearchDataFilter
+from .research_strategies import get_strategy_for_mode
 
 
 class ResearchService:
@@ -44,7 +48,6 @@ class ResearchService:
         Includes intelligent caching for exact keyword matches.
         """
         try:
-            from services.llm_providers.gemini_grounded_provider import GeminiGroundedProvider
             from services.cache.research_cache import research_cache
             
             topic = request.topic or ", ".join(request.keywords)
@@ -79,62 +82,104 @@ class ResearchService:
             
             # Cache miss - proceed with API call
             logger.info(f"Cache miss - making API call for keywords: {request.keywords}")
-            blog_writer_logger.log_operation_start("gemini_api_call", api_name="gemini_grounded", operation="research")
-            gemini = GeminiGroundedProvider()
+            blog_writer_logger.log_operation_start("research_api_call", api_name="research", operation="research")
 
-            # Single comprehensive research prompt - Gemini handles Google Search automatically
-            research_prompt = f"""
-            Research the topic "{topic}" in the {industry} industry for {target_audience} audience. Provide a comprehensive analysis including:
-
-            1. Current trends and insights (2024-2025)
-            2. Key statistics and data points with sources
-            3. Industry expert opinions and quotes
-            4. Recent developments and news
-            5. Market analysis and forecasts
-            6. Best practices and case studies
-            7. Keyword analysis: primary, secondary, and long-tail opportunities
-            8. Competitor analysis: top players and content gaps
-            9. Content angle suggestions: 5 compelling angles for blog posts
-
-            Focus on factual, up-to-date information from credible sources.
-            Include specific data points, percentages, and recent developments.
-            Structure your response with clear sections for each analysis area.
-            """
+            # Determine research mode and get appropriate strategy
+            research_mode = request.research_mode or ResearchMode.BASIC
+            config = request.config or ResearchConfig(mode=research_mode, provider=ResearchProvider.GOOGLE)
+            strategy = get_strategy_for_mode(research_mode)
             
-            # Single Gemini call with native Google Search grounding - no fallbacks
-            # Validation is handled inside generate_grounded_content when validate_subsequent_operations=True
-            import time
-            api_start_time = time.time()
-            gemini_result = await gemini.generate_grounded_content(
-                prompt=research_prompt,
-                content_type="research",
-                max_tokens=2000,
-                user_id=user_id,
-                validate_subsequent_operations=True  # Validates Google Grounding + 3 LLM calls
-            )
-            api_duration_ms = (time.time() - api_start_time) * 1000
+            logger.info(f"Research: mode={research_mode.value}, provider={config.provider.value}")
             
-            # Log API call performance
-            blog_writer_logger.log_api_call(
-                "gemini_grounded",
-                "generate_grounded_content",
-                api_duration_ms,
-                token_usage=gemini_result.get("token_usage", {}),
-                content_length=len(gemini_result.get("content", ""))
-            )
+            # Build research prompt based on strategy
+            research_prompt = strategy.build_research_prompt(topic, industry, target_audience, config)
             
-            # Extract sources from grounding metadata
-            sources = self._extract_sources_from_grounding(gemini_result)
+            # Route to appropriate provider
+            if config.provider == ResearchProvider.EXA:
+                # Exa research workflow
+                from .exa_provider import ExaResearchProvider
+                from services.subscription.preflight_validator import validate_exa_research_operations
+                from services.database import get_db
+                from services.subscription import PricingService
+                import os
+                import time
+                
+                # Pre-flight validation
+                db_val = next(get_db())
+                try:
+                    pricing_service = PricingService(db_val)
+                    gpt_provider = os.getenv("GPT_PROVIDER", "google")
+                    validate_exa_research_operations(pricing_service, user_id, gpt_provider)
+                finally:
+                    db_val.close()
+                
+                # Execute Exa search
+                api_start_time = time.time()
+                try:
+                    exa_provider = ExaResearchProvider()
+                    raw_result = await exa_provider.search(
+                        research_prompt, topic, industry, target_audience, config, user_id
+                    )
+                    api_duration_ms = (time.time() - api_start_time) * 1000
+                    
+                    # Track usage
+                    cost = raw_result.get('cost', {}).get('total', 0.005) if isinstance(raw_result.get('cost'), dict) else 0.005
+                    exa_provider.track_exa_usage(user_id, cost)
+                    
+                    # Log API call performance
+                    blog_writer_logger.log_api_call(
+                        "exa_search",
+                        "search_and_contents",
+                        api_duration_ms,
+                        token_usage={},
+                        content_length=len(raw_result.get('content', ''))
+                    )
+                    
+                    # Extract content for downstream analysis
+                    content = raw_result.get('content', '')
+                    sources = raw_result.get('sources', [])
+                    search_widget = ""  # Exa doesn't provide search widgets
+                    search_queries = raw_result.get('search_queries', [])
+                    grounding_metadata = None  # Exa doesn't provide grounding metadata
+                    
+                except RuntimeError as e:
+                    if "EXA_API_KEY not configured" in str(e):
+                        logger.warning("Exa not configured, falling back to Google")
+                        config.provider = ResearchProvider.GOOGLE
+                        # Continue to Google flow below
+                        raw_result = None
+                    else:
+                        raise
+                
+            if config.provider != ResearchProvider.EXA:
+                # Google research (existing flow) or fallback from Exa
+                from .google_provider import GoogleResearchProvider
+                import time
+                
+                api_start_time = time.time()
+                google_provider = GoogleResearchProvider()
+                gemini_result = await google_provider.search(
+                    research_prompt, topic, industry, target_audience, config, user_id
+                )
+                api_duration_ms = (time.time() - api_start_time) * 1000
+                
+                # Log API call performance
+                blog_writer_logger.log_api_call(
+                    "gemini_grounded",
+                    "generate_grounded_content",
+                    api_duration_ms,
+                    token_usage=gemini_result.get("token_usage", {}),
+                    content_length=len(gemini_result.get("content", ""))
+                )
+                
+                # Extract sources and content
+                sources = self._extract_sources_from_grounding(gemini_result)
+                content = gemini_result.get("content", "")
+                search_widget = gemini_result.get("search_widget", "") or ""
+                search_queries = gemini_result.get("search_queries", []) or []
+                grounding_metadata = self._extract_grounding_metadata(gemini_result)
             
-            # Extract grounding metadata for detailed UI display
-            grounding_metadata = self._extract_grounding_metadata(gemini_result)
-            
-            # Extract search widget and queries for UI display
-            search_widget = gemini_result.get("search_widget", "") or ""
-            search_queries = gemini_result.get("search_queries", []) or []
-            
-            # Parse the comprehensive response for different analysis components
-            content = gemini_result.get("content", "")
+            # Continue with common analysis (same for both providers)
             keyword_analysis = self.keyword_analyzer.analyze(content, request.keywords, user_id=user_id)
             competitor_analysis = self.competitor_analyzer.analyze(content, user_id=user_id)
             suggested_angles = self.content_angle_generator.generate(content, topic, industry, user_id=user_id)
@@ -261,7 +306,6 @@ class ResearchService:
         Research method with progress updates for real-time feedback.
         """
         try:
-            from services.llm_providers.gemini_grounded_provider import GeminiGroundedProvider
             from services.cache.research_cache import research_cache
             from services.cache.persistent_research_cache import persistent_research_cache
             from api.blog_writer.task_manager import task_manager
@@ -293,66 +337,100 @@ class ResearchService:
                 logger.info(f"Returning cached research result for keywords: {request.keywords}")
                 return BlogResearchResponse(**cached_result)
             
-            # User ID validation (validation logic is now in Google Grounding provider)
+            # User ID validation
             if not user_id:
                 await task_manager.update_progress(task_id, "❌ Error: User ID is required for research operation")
                 raise ValueError("user_id is required for research operation. Please provide Clerk user ID.")
             
-            # Cache miss - proceed with API call
-            await task_manager.update_progress(task_id, "🌐 Cache miss - connecting to Google Search grounding...")
-            logger.info(f"Cache miss - making API call for keywords: {request.keywords}")
-            gemini = GeminiGroundedProvider()
-
-            # Single comprehensive research prompt - Gemini handles Google Search automatically
-            research_prompt = f"""
-            Research the topic "{topic}" in the {industry} industry for {target_audience} audience. Provide a comprehensive analysis including:
-
-            1. Current trends and insights (2024-2025)
-            2. Key statistics and data points with sources
-            3. Industry expert opinions and quotes
-            4. Recent developments and news
-            5. Market analysis and forecasts
-            6. Best practices and case studies
-            7. Keyword analysis: primary, secondary, and long-tail opportunities
-            8. Competitor analysis: top players and content gaps
-            9. Content angle suggestions: 5 compelling angles for blog posts
-
-            Focus on factual, up-to-date information from credible sources.
-            Include specific data points, percentages, and recent developments.
-            Structure your response with clear sections for each analysis area.
-            """
+            # Determine research mode and get appropriate strategy
+            research_mode = request.research_mode or ResearchMode.BASIC
+            config = request.config or ResearchConfig(mode=research_mode, provider=ResearchProvider.GOOGLE)
+            strategy = get_strategy_for_mode(research_mode)
             
-            await task_manager.update_progress(task_id, "🤖 Making AI request to Gemini with Google Search grounding...")
-            # Single Gemini call with native Google Search grounding - no fallbacks
-            # Validation is handled inside generate_grounded_content when validate_subsequent_operations=True
-            try:
-                gemini_result = await gemini.generate_grounded_content(
-                    prompt=research_prompt,
-                    content_type="research",
-                    max_tokens=2000,
-                    user_id=user_id,
-                    validate_subsequent_operations=True  # Validates Google Grounding + 3 LLM calls
-                )
-            except HTTPException as http_error:
-                # Re-raise HTTPException so it can be properly handled by task manager
-                logger.error(f"Subscription limit exceeded for research: {http_error.detail}")
-                await task_manager.update_progress(task_id, f"❌ Subscription limit exceeded: {http_error.detail.get('message', str(http_error.detail)) if isinstance(http_error.detail, dict) else str(http_error.detail)}")
-                raise  # Re-raise HTTPException to preserve status code and error details
+            logger.info(f"Research: mode={research_mode.value}, provider={config.provider.value}")
             
-            await task_manager.update_progress(task_id, "📊 Processing research results and extracting insights...")
-            # Extract sources from grounding metadata
-            sources = self._extract_sources_from_grounding(gemini_result)
+            # Build research prompt based on strategy
+            research_prompt = strategy.build_research_prompt(topic, industry, target_audience, config)
             
-            # Extract grounding metadata for detailed UI display
-            grounding_metadata = self._extract_grounding_metadata(gemini_result)
-            
-            # Extract search widget and queries for UI display
-            search_widget = gemini_result.get("search_widget", "") or ""
-            search_queries = gemini_result.get("search_queries", []) or []
+            # Route to appropriate provider
+            if config.provider == ResearchProvider.EXA:
+                # Exa research workflow
+                from .exa_provider import ExaResearchProvider
+                from services.subscription.preflight_validator import validate_exa_research_operations
+                from services.database import get_db
+                from services.subscription import PricingService
+                import os
+                
+                await task_manager.update_progress(task_id, "🌐 Connecting to Exa neural search...")
+                
+                # Pre-flight validation
+                db_val = next(get_db())
+                try:
+                    pricing_service = PricingService(db_val)
+                    gpt_provider = os.getenv("GPT_PROVIDER", "google")
+                    validate_exa_research_operations(pricing_service, user_id, gpt_provider)
+                except HTTPException as http_error:
+                    logger.error(f"Subscription limit exceeded for Exa research: {http_error.detail}")
+                    await task_manager.update_progress(task_id, f"❌ Subscription limit exceeded: {http_error.detail.get('message', str(http_error.detail)) if isinstance(http_error.detail, dict) else str(http_error.detail)}")
+                    raise
+                finally:
+                    db_val.close()
+                
+                # Execute Exa search
+                await task_manager.update_progress(task_id, "🤖 Executing Exa neural search...")
+                try:
+                    exa_provider = ExaResearchProvider()
+                    raw_result = await exa_provider.search(
+                        research_prompt, topic, industry, target_audience, config, user_id
+                    )
+                    
+                    # Track usage
+                    cost = raw_result.get('cost', {}).get('total', 0.005) if isinstance(raw_result.get('cost'), dict) else 0.005
+                    exa_provider.track_exa_usage(user_id, cost)
+                    
+                    # Extract content for downstream analysis
+                    content = raw_result.get('content', '')
+                    sources = raw_result.get('sources', [])
+                    search_widget = ""  # Exa doesn't provide search widgets
+                    search_queries = raw_result.get('search_queries', [])
+                    grounding_metadata = None  # Exa doesn't provide grounding metadata
+                    
+                except RuntimeError as e:
+                    if "EXA_API_KEY not configured" in str(e):
+                        logger.warning("Exa not configured, falling back to Google")
+                        await task_manager.update_progress(task_id, "⚠️ Exa not configured, falling back to Google Search")
+                        config.provider = ResearchProvider.GOOGLE
+                        # Continue to Google flow below
+                    else:
+                        raise
+                
+            if config.provider != ResearchProvider.EXA:
+                # Google research (existing flow)
+                from .google_provider import GoogleResearchProvider
+                
+                await task_manager.update_progress(task_id, "🌐 Connecting to Google Search grounding...")
+                google_provider = GoogleResearchProvider()
+                
+                await task_manager.update_progress(task_id, "🤖 Making AI request to Gemini with Google Search grounding...")
+                try:
+                    gemini_result = await google_provider.search(
+                        research_prompt, topic, industry, target_audience, config, user_id
+                    )
+                except HTTPException as http_error:
+                    logger.error(f"Subscription limit exceeded for Google research: {http_error.detail}")
+                    await task_manager.update_progress(task_id, f"❌ Subscription limit exceeded: {http_error.detail.get('message', str(http_error.detail)) if isinstance(http_error.detail, dict) else str(http_error.detail)}")
+                    raise
+                
+                await task_manager.update_progress(task_id, "📊 Processing research results and extracting insights...")
+                # Extract sources and content
+                sources = self._extract_sources_from_grounding(gemini_result)
+                content = gemini_result.get("content", "")
+                search_widget = gemini_result.get("search_widget", "") or ""
+                search_queries = gemini_result.get("search_queries", []) or []
+                grounding_metadata = self._extract_grounding_metadata(gemini_result)
             
+            # Continue with common analysis (same for both providers)
             await task_manager.update_progress(task_id, "🔍 Analyzing keywords and content angles...")
-            # Parse the comprehensive response for different analysis components
-            content = gemini_result.get("content", "")
             keyword_analysis = self.keyword_analyzer.analyze(content, request.keywords, user_id=user_id)
             competitor_analysis = self.competitor_analyzer.analyze(content, user_id=user_id)
             suggested_angles = self.content_angle_generator.generate(content, topic, industry, user_id=user_id)
diff --git a/backend/services/blog_writer/research/research_strategies.py b/backend/services/blog_writer/research/research_strategies.py
new file mode 100644
index 00000000..5de20e34
--- /dev/null
+++ b/backend/services/blog_writer/research/research_strategies.py
@@ -0,0 +1,234 @@
+"""
+Research Strategy Pattern Implementation
+
+Different strategies for executing research based on depth and focus.
+"""
+
+from abc import ABC, abstractmethod
+from typing import Dict, Any
+from loguru import logger
+
+from models.blog_models import BlogResearchRequest, ResearchMode, ResearchConfig
+from .keyword_analyzer import KeywordAnalyzer
+from .competitor_analyzer import CompetitorAnalyzer
+from .content_angle_generator import ContentAngleGenerator
+
+
+class ResearchStrategy(ABC):
+    """Base class for research strategies."""
+    
+    def __init__(self):
+        self.keyword_analyzer = KeywordAnalyzer()
+        self.competitor_analyzer = CompetitorAnalyzer()
+        self.content_angle_generator = ContentAngleGenerator()
+    
+    @abstractmethod
+    def build_research_prompt(
+        self, 
+        topic: str, 
+        industry: str, 
+        target_audience: str,
+        config: ResearchConfig
+    ) -> str:
+        """Build the research prompt for the strategy."""
+        pass
+    
+    @abstractmethod
+    def get_mode(self) -> ResearchMode:
+        """Return the research mode this strategy handles."""
+        pass
+
+
+class BasicResearchStrategy(ResearchStrategy):
+    """Basic research strategy - keyword focused, minimal analysis."""
+    
+    def get_mode(self) -> ResearchMode:
+        return ResearchMode.BASIC
+    
+    def build_research_prompt(
+        self,
+        topic: str,
+        industry: str,
+        target_audience: str,
+        config: ResearchConfig
+    ) -> str:
+        """Build basic research prompt focused on keywords and quick insights."""
+        prompt = f"""You are a professional blog content strategist researching for a {industry} blog targeting {target_audience}.
+
+Research Topic: "{topic}"
+
+Provide analysis in this EXACT format:
+
+## CURRENT TRENDS (2024-2025)
+- [Trend 1 with specific data and source URL]
+- [Trend 2 with specific data and source URL]
+- [Trend 3 with specific data and source URL]
+
+## KEY STATISTICS
+- [Statistic 1: specific number/percentage with source URL]
+- [Statistic 2: specific number/percentage with source URL]
+- [Statistic 3: specific number/percentage with source URL]
+- [Statistic 4: specific number/percentage with source URL]
+- [Statistic 5: specific number/percentage with source URL]
+
+## PRIMARY KEYWORDS
+1. "{topic}" (main keyword)
+2. [Variation 1]
+3. [Variation 2]
+
+## SECONDARY KEYWORDS
+[5 related keywords for blog content]
+
+## CONTENT ANGLES (Top 5)
+1. [Angle 1: specific unique approach]
+2. [Angle 2: specific unique approach]
+3. [Angle 3: specific unique approach]
+4. [Angle 4: specific unique approach]
+5. [Angle 5: specific unique approach]
+
+REQUIREMENTS:
+- Cite EVERY claim with authoritative source URLs
+- Use 2024-2025 data when available
+- Include specific numbers, dates, examples
+- Focus on actionable blog insights for {target_audience}"""
+        return prompt.strip()
+
+
+class ComprehensiveResearchStrategy(ResearchStrategy):
+    """Comprehensive research strategy - full analysis with all components."""
+    
+    def get_mode(self) -> ResearchMode:
+        return ResearchMode.COMPREHENSIVE
+    
+    def build_research_prompt(
+        self,
+        topic: str,
+        industry: str,
+        target_audience: str,
+        config: ResearchConfig
+    ) -> str:
+        """Build comprehensive research prompt with all analysis components."""
+        date_filter = f"\nDate Focus: {config.date_range.value.replace('_', ' ')}" if config.date_range else ""
+        source_filter = f"\nPriority Sources: {', '.join([s.value for s in config.source_types])}" if config.source_types else ""
+        
+        prompt = f"""You are a senior blog content strategist conducting comprehensive research for a {industry} blog targeting {target_audience}.
+
+Research Topic: "{topic}"{date_filter}{source_filter}
+
+Provide COMPLETE analysis in this EXACT format:
+
+## TRENDS AND INSIGHTS (2024-2025)
+[5-7 trends with specific data, numbers, and source URLs]
+
+## KEY STATISTICS
+[7-10 statistics with exact numbers, percentages, dates, and source URLs]
+
+## EXPERT OPINIONS
+[4-5 expert quotes with full attribution and source URLs]
+
+## RECENT DEVELOPMENTS
+[5-7 recent news/developments with dates and source URLs]
+
+## MARKET ANALYSIS
+[3-5 market insights with data points and source URLs]
+
+## BEST PRACTICES & CASE STUDIES
+[3-5 examples with specific outcomes/metrics and source URLs]
+
+## KEYWORD ANALYSIS
+Primary Keywords: [3 main variations]
+Secondary Keywords: [7-10 related keywords]
+Long-Tail Opportunities: [5-7 specific search phrases]
+
+## COMPETITOR ANALYSIS
+Top Competitors: [5 competitors with brief descriptions]
+Content Gaps: [5 topics competitors are missing]
+Competitive Advantages: [5 unique angles we can own]
+
+## CONTENT ANGLES (Exactly 5)
+1. [Unique angle with reasoning and target benefit]
+2. [Unique angle with reasoning and target benefit]
+3. [Unique angle with reasoning and target benefit]
+4. [Unique angle with reasoning and target benefit]
+5. [Unique angle with reasoning and target benefit]
+
+VERIFICATION REQUIREMENTS:
+- Minimum 2 authoritative sources per major claim
+- Prioritize: Industry publications > Research papers > News > Blogs
+- 2024-2025 data strongly preferred
+- All numbers must include context (timeframe, sample size, methodology)
+- Every recommendation must be actionable for {target_audience}"""
+        return prompt.strip()
+
+
+class TargetedResearchStrategy(ResearchStrategy):
+    """Targeted research strategy - focused on specific aspects."""
+    
+    def get_mode(self) -> ResearchMode:
+        return ResearchMode.TARGETED
+    
+    def build_research_prompt(
+        self,
+        topic: str,
+        industry: str,
+        target_audience: str,
+        config: ResearchConfig
+    ) -> str:
+        """Build targeted research prompt based on config preferences."""
+        sections = []
+        
+        if config.include_trends:
+            sections.append("""## CURRENT TRENDS
+[3-5 trends with data and source URLs]""")
+        
+        if config.include_statistics:
+            sections.append("""## KEY STATISTICS
+[5-7 statistics with numbers and source URLs]""")
+        
+        if config.include_expert_quotes:
+            sections.append("""## EXPERT OPINIONS
+[3-4 expert quotes with attribution and source URLs]""")
+        
+        if config.include_competitors:
+            sections.append("""## COMPETITOR ANALYSIS
+Top Competitors: [3-5]
+Content Gaps: [3-5]""")
+        
+        # Always include keywords and angles
+        sections.append("""## KEYWORD ANALYSIS
+Primary: [2-3 variations]
+Secondary: [5-7 keywords]
+Long-Tail: [3-5 phrases]""")
+        
+        sections.append("""## CONTENT ANGLES (3-5)
+[Unique blog angles with reasoning]""")
+        
+        sections_str = "\n\n".join(sections)
+        
+        prompt = f"""You are a blog content strategist conducting targeted research for a {industry} blog targeting {target_audience}.
+
+Research Topic: "{topic}"
+
+Provide focused analysis in this EXACT format:
+
+{sections_str}
+
+REQUIREMENTS:
+- Cite all claims with authoritative source URLs
+- Include specific numbers, dates, examples
+- Focus on actionable insights for {target_audience}
+- Use 2024-2025 data when available"""
+        return prompt.strip()
+
+
+def get_strategy_for_mode(mode: ResearchMode) -> ResearchStrategy:
+    """Factory function to get the appropriate strategy for a mode."""
+    strategy_map = {
+        ResearchMode.BASIC: BasicResearchStrategy,
+        ResearchMode.COMPREHENSIVE: ComprehensiveResearchStrategy,
+        ResearchMode.TARGETED: TargetedResearchStrategy,
+    }
+    
+    strategy_class = strategy_map.get(mode, BasicResearchStrategy)
+    return strategy_class()
+
diff --git a/backend/services/integrations/wix/__init__.py b/backend/services/integrations/wix/__init__.py
index 132171a6..67c5a433 100644
--- a/backend/services/integrations/wix/__init__.py
+++ b/backend/services/integrations/wix/__init__.py
@@ -2,4 +2,14 @@
 Wix integration modular services package.
 """
 
+from services.integrations.wix.seo import build_seo_data
+from services.integrations.wix.ricos_converter import markdown_to_html, convert_via_wix_api
+from services.integrations.wix.blog_publisher import create_blog_post
+
+__all__ = [
+    'build_seo_data',
+    'markdown_to_html',
+    'convert_via_wix_api',
+    'create_blog_post',
+]
 
diff --git a/backend/services/integrations/wix/blog.py b/backend/services/integrations/wix/blog.py
index 305158ce..6476e05a 100644
--- a/backend/services/integrations/wix/blog.py
+++ b/backend/services/integrations/wix/blog.py
@@ -20,6 +20,40 @@ class WixBlogService:
         return h
 
     def create_draft_post(self, access_token: str, payload: Dict[str, Any], extra_headers: Optional[Dict[str, str]] = None) -> Dict[str, Any]:
+        # Log the exact payload being sent for debugging
+        import json
+        logger.warning(f"📤 Sending to Wix Blog API:")
+        logger.warning(f"  Endpoint: {self.base_url}/blog/v3/draft-posts")
+        logger.warning(f"  Payload top-level keys: {list(payload.keys())}")
+        if 'draftPost' in payload:
+            dp = payload['draftPost']
+            logger.warning(f"  draftPost keys: {list(dp.keys())}")
+            if 'richContent' in dp:
+                rc = dp['richContent']
+                logger.warning(f"  richContent keys: {list(rc.keys()) if isinstance(rc, dict) else 'N/A'}")
+                if isinstance(rc, dict) and 'nodes' in rc:
+                    nodes = rc['nodes']
+                    logger.warning(f"  richContent.nodes count: {len(nodes) if isinstance(nodes, list) else 'N/A'}")
+                    # Inspect first LIST_ITEM node if any
+                    for i, node in enumerate(nodes[:10]):
+                        if isinstance(node, dict) and node.get('type') == 'LIST_ITEM':
+                            logger.warning(f"  Found LIST_ITEM at index {i}:")
+                            logger.warning(f"    Keys: {list(node.keys())}")
+                            logger.warning(f"    Has listItemData: {'listItemData' in node}")
+                            if 'listItemData' in node:
+                                logger.warning(f"    listItemData type: {type(node['listItemData'])}, value: {node['listItemData']}")
+                            if 'nodes' in node:
+                                nested = node['nodes']
+                                logger.warning(f"    Nested nodes count: {len(nested) if isinstance(nested, list) else 'N/A'}")
+                                for j, n_node in enumerate(nested[:3]):
+                                    if isinstance(n_node, dict):
+                                        logger.warning(f"      Nested node {j}: type={n_node.get('type')}, keys={list(n_node.keys())}")
+                                        if n_node.get('type') == 'PARAGRAPH' and 'paragraphData' in n_node:
+                                            logger.warning(f"        paragraphData type: {type(n_node['paragraphData'])}, value: {n_node['paragraphData']}")
+                            break  # Only inspect first LIST_ITEM
+        
+        logger.warning(f"  Full Payload JSON (first 8000 chars):\n{json.dumps(payload, indent=2, ensure_ascii=False)[:8000]}...")
+        
         response = requests.post(f"{self.base_url}/blog/v3/draft-posts", headers=self.headers(access_token, extra_headers), json=payload)
         response.raise_for_status()
         return response.json()
diff --git a/backend/services/integrations/wix/blog_publisher.py b/backend/services/integrations/wix/blog_publisher.py
new file mode 100644
index 00000000..39c2d916
--- /dev/null
+++ b/backend/services/integrations/wix/blog_publisher.py
@@ -0,0 +1,716 @@
+"""
+Blog Post Publisher for Wix
+
+Handles blog post creation, validation, and publishing to Wix.
+"""
+
+import json
+import uuid
+import requests
+import jwt
+from typing import Dict, Any, Optional, List
+from loguru import logger
+from services.integrations.wix.blog import WixBlogService
+from services.integrations.wix.content import convert_content_to_ricos
+from services.integrations.wix.ricos_converter import convert_via_wix_api
+from services.integrations.wix.seo import build_seo_data
+
+
+def validate_ricos_content(ricos_content: Dict[str, Any]) -> Dict[str, Any]:
+    """
+    Validate and normalize Ricos document structure.
+    
+    Args:
+        ricos_content: Ricos document dict
+        
+    Returns:
+        Validated and normalized Ricos document
+    """
+    # Validate Ricos document structure before using
+    if not ricos_content or not isinstance(ricos_content, dict):
+        logger.error("Invalid Ricos content - not a dict")
+        raise ValueError("Failed to convert content to valid Ricos format")
+    
+    if 'type' not in ricos_content:
+        ricos_content['type'] = 'DOCUMENT'
+        logger.debug("Added missing richContent type 'DOCUMENT'")
+    if ricos_content.get('type') != 'DOCUMENT':
+        logger.warning(f"richContent type expected 'DOCUMENT', got {ricos_content.get('type')}, correcting")
+        ricos_content['type'] = 'DOCUMENT'
+
+    if 'id' not in ricos_content or not isinstance(ricos_content.get('id'), str):
+        ricos_content['id'] = str(uuid.uuid4())
+        logger.debug("Added missing richContent id")
+
+    if 'nodes' not in ricos_content:
+        logger.warning("Ricos document missing 'nodes' field, adding empty nodes array")
+        ricos_content['nodes'] = []
+    
+    logger.debug(f"Ricos document structure: nodes={len(ricos_content.get('nodes', []))}")
+
+    # Validate richContent is a proper object with nodes array
+    # Per Wix API: richContent must be a RichContent object with nodes array
+    if not isinstance(ricos_content, dict):
+        raise ValueError(f"richContent must be a dict object, got {type(ricos_content)}")
+    
+    # Ensure nodes array exists and is valid
+    if 'nodes' not in ricos_content:
+        logger.warning("richContent missing 'nodes', adding empty array")
+        ricos_content['nodes'] = []
+    
+    if not isinstance(ricos_content['nodes'], list):
+        raise ValueError(f"richContent.nodes must be a list, got {type(ricos_content['nodes'])}")
+    
+    # Recursive function to validate and fix nodes at any depth
+    def validate_node_recursive(node: Dict[str, Any], path: str = "root") -> None:
+        """
+        Recursively validate a node and all its nested children, ensuring:
+        1. All required data fields exist for each node type
+        2. All 'nodes' arrays are proper lists
+        3. No None values in critical fields
+        """
+        if not isinstance(node, dict):
+            logger.error(f"{path}: Node is not a dict: {type(node)}")
+            return
+        
+        # Ensure type and id exist
+        if 'type' not in node:
+            logger.error(f"{path}: Missing 'type' field - REQUIRED")
+            node['type'] = 'PARAGRAPH'  # Default fallback
+        if 'id' not in node:
+            node['id'] = str(uuid.uuid4())
+            logger.debug(f"{path}: Added missing 'id'")
+        
+        node_type = node.get('type')
+        
+        # CRITICAL: Per Wix API schema, data fields like paragraphData, bulletedListData, etc.
+        # are OPTIONAL and should be OMITTED entirely when empty, not included as {}
+        # Only validate fields that have required properties
+        
+        # Special handling: Remove listItemData if it exists (not in Wix API schema)
+        if node_type == 'LIST_ITEM' and 'listItemData' in node:
+            logger.debug(f"{path}: Removing incorrect listItemData field from LIST_ITEM")
+            del node['listItemData']
+        
+        # Only validate HEADING nodes - they require headingData with level property
+        if node_type == 'HEADING':
+            if 'headingData' not in node or not isinstance(node.get('headingData'), dict):
+                logger.warning(f"{path} (HEADING): Missing headingData, adding default level 1")
+                node['headingData'] = {'level': 1}
+            elif 'level' not in node['headingData']:
+                logger.warning(f"{path} (HEADING): Missing level in headingData, adding default")
+                node['headingData']['level'] = 1
+        
+        # TEXT nodes must have textData
+        if node_type == 'TEXT':
+            if 'textData' not in node or not isinstance(node.get('textData'), dict):
+                logger.error(f"{path} (TEXT): Missing/invalid textData - node will be problematic")
+                node['textData'] = {'text': '', 'decorations': []}
+        
+        # LINK and IMAGE nodes must have their data fields
+        if node_type == 'LINK' and ('linkData' not in node or not isinstance(node.get('linkData'), dict)):
+            logger.error(f"{path} (LINK): Missing/invalid linkData - node will be problematic")
+        if node_type == 'IMAGE' and ('imageData' not in node or not isinstance(node.get('imageData'), dict)):
+            logger.error(f"{path} (IMAGE): Missing/invalid imageData - node will be problematic")
+        
+        # Remove None values from any data fields that exist (Wix API rejects None)
+        for data_field in ['headingData', 'paragraphData', 'blockquoteData', 'bulletedListData', 
+                          'orderedListData', 'textData', 'linkData', 'imageData']:
+            if data_field in node and isinstance(node[data_field], dict):
+                data_value = node[data_field]
+                keys_to_remove = [k for k, v in data_value.items() if v is None]
+                if keys_to_remove:
+                    logger.debug(f"{path} ({node_type}): Removing None values from {data_field}: {keys_to_remove}")
+                    for key in keys_to_remove:
+                        del data_value[key]
+        
+        # Ensure 'nodes' field exists for container nodes
+        container_types = ['HEADING', 'PARAGRAPH', 'BLOCKQUOTE', 'LIST_ITEM', 'LINK', 
+                          'BULLETED_LIST', 'ORDERED_LIST']
+        if node_type in container_types:
+            if 'nodes' not in node:
+                logger.warning(f"{path} ({node_type}): Missing 'nodes' field, adding empty array")
+                node['nodes'] = []
+            elif not isinstance(node['nodes'], list):
+                logger.error(f"{path} ({node_type}): Invalid 'nodes' field (not a list), fixing")
+                node['nodes'] = []
+            
+            # Recursively validate all nested nodes
+            for nested_idx, nested_node in enumerate(node['nodes']):
+                nested_path = f"{path}.nodes[{nested_idx}]"
+                validate_node_recursive(nested_node, nested_path)
+    
+    # Validate all top-level nodes recursively
+    for idx, node in enumerate(ricos_content['nodes']):
+        validate_node_recursive(node, f"nodes[{idx}]")
+    
+    # Ensure documentStyle exists and is a dict (required by Wix API when provided)
+    if 'metadata' not in ricos_content or not isinstance(ricos_content.get('metadata'), dict):
+        ricos_content['metadata'] = {'version': 1, 'id': str(uuid.uuid4())}
+        logger.debug("Added default metadata to richContent")
+    else:
+        ricos_content['metadata'].setdefault('version', 1)
+        ricos_content['metadata'].setdefault('id', str(uuid.uuid4()))
+    
+    if 'documentStyle' not in ricos_content or not isinstance(ricos_content.get('documentStyle'), dict):
+        ricos_content['documentStyle'] = {
+            'paragraph': {
+                'decorations': [],
+                'nodeStyle': {},
+                'lineHeight': '1.5'
+            }
+        }
+        logger.debug("Added default documentStyle to richContent")
+    
+    logger.debug(f"✅ Validated richContent: {len(ricos_content['nodes'])} nodes, has_metadata={bool(ricos_content.get('metadata'))}, has_documentStyle={bool(ricos_content.get('documentStyle'))}")
+    
+    return ricos_content
+
+
+def validate_payload_no_none(obj, path=""):
+    """Recursively validate that no None values exist in the payload"""
+    if obj is None:
+        raise ValueError(f"Found None value at path: {path}")
+    if isinstance(obj, dict):
+        for key, value in obj.items():
+            validate_payload_no_none(value, f"{path}.{key}" if path else key)
+    elif isinstance(obj, list):
+        for idx, item in enumerate(obj):
+            validate_payload_no_none(item, f"{path}[{idx}]" if path else f"[{idx}]")
+
+
+def create_blog_post(
+    blog_service: WixBlogService,
+    access_token: str,
+    title: str,
+    content: str,
+    member_id: str,
+    cover_image_url: str = None,
+    category_ids: List[str] = None,
+    tag_ids: List[str] = None,
+    publish: bool = True,
+    seo_metadata: Dict[str, Any] = None,
+    import_image_func = None,
+    lookup_categories_func = None,
+    lookup_tags_func = None,
+    base_url: str = 'https://www.wixapis.com'
+) -> Dict[str, Any]:
+    """
+    Create and optionally publish a blog post on Wix
+    
+    Args:
+        blog_service: WixBlogService instance
+        access_token: Valid access token
+        title: Blog post title
+        content: Blog post content (markdown)
+        member_id: Required for third-party apps - the member ID of the post author
+        cover_image_url: Optional cover image URL
+        category_ids: Optional list of category IDs or names
+        tag_ids: Optional list of tag IDs or names
+        publish: Whether to publish immediately or save as draft
+        seo_metadata: Optional SEO metadata dict
+        import_image_func: Function to import images (optional)
+        lookup_categories_func: Function to lookup/create categories (optional)
+        lookup_tags_func: Function to lookup/create tags (optional)
+        base_url: Wix API base URL
+        
+    Returns:
+        Created blog post information
+    """
+    if not member_id:
+        raise ValueError("memberId is required for third-party apps creating blog posts")
+    
+    headers = {
+        'Authorization': f'Bearer {access_token}',
+        'Content-Type': 'application/json'
+    }
+    
+    # Build valid Ricos rich content
+    # Ensure content is not empty
+    if not content or not content.strip():
+        content = "This is a post from ALwrity."
+        logger.warning("⚠️ Content was empty, using default text")
+    
+    # Try Wix API first (more reliable), fall back to custom parser
+    ricos_content = None
+    try:
+        logger.warning("🔄 Attempting to convert markdown to Ricos via Wix API...")
+        ricos_content = convert_via_wix_api(content, access_token, base_url)
+        logger.warning(f"✅ Wix API conversion successful. Ricos document has {len(ricos_content.get('nodes', []))} nodes")
+    except Exception as e:
+        logger.warning(f"⚠️ Wix Ricos API conversion failed: {e}. Falling back to custom parser...")
+        # Fall back to custom parser
+        ricos_content = convert_content_to_ricos(content, None)
+        logger.warning(f"✅ Custom parser conversion complete. Ricos document has {len(ricos_content.get('nodes', []))} nodes")
+    
+    # Validate Ricos content
+    ricos_content = validate_ricos_content(ricos_content)
+    
+    # Minimal payload per Wix docs: title, memberId, and richContent
+    # CRITICAL: Only include fields that have valid values (no None, no empty strings for required fields)
+    blog_data = {
+        'draftPost': {
+            'title': str(title).strip() if title else "Untitled",
+            'memberId': str(member_id).strip(),  # Required for third-party apps (validated above)
+            'richContent': ricos_content,  # Must be a valid Ricos document object
+        },
+        'publish': bool(publish),
+        'fieldsets': ['URL']  # Simplified fieldsets
+    }
+    
+    # Add excerpt only if content exists and is not empty (avoid None or empty strings)
+    excerpt = (content or '').strip()[:200] if content else None
+    if excerpt and len(excerpt) > 0:
+        blog_data['draftPost']['excerpt'] = str(excerpt)
+    
+    # Add cover image if provided
+    if cover_image_url and import_image_func:
+        try:
+            media_id = import_image_func(access_token, cover_image_url, f'Cover: {title}')
+            # Ensure media_id is a string and not None
+            if media_id and isinstance(media_id, str):
+                blog_data['draftPost']['media'] = {
+                    'wixMedia': {
+                        'image': {'id': str(media_id).strip()}
+                    },
+                    'displayed': True,
+                    'custom': True
+                }
+            else:
+                logger.warning(f"Invalid media_id type or value: {type(media_id)}, skipping media")
+        except Exception as e:
+            logger.warning(f"Failed to import cover image: {e}")
+    
+    # Handle categories - can be either IDs (list of strings) or names (for lookup)
+    category_ids_to_use = None
+    if category_ids:
+        # Check if these are IDs (UUIDs) or names
+        if isinstance(category_ids, list) and len(category_ids) > 0:
+            # Assume IDs if first item looks like UUID (has hyphens and is long)
+            first_item = str(category_ids[0])
+            if '-' in first_item and len(first_item) > 30:
+                category_ids_to_use = category_ids
+            elif lookup_categories_func:
+                # These are names, need to lookup/create
+                extra_headers = {}
+                if 'wix-site-id' in headers:
+                    extra_headers['wix-site-id'] = headers['wix-site-id']
+                category_ids_to_use = lookup_categories_func(
+                    access_token, category_ids, extra_headers if extra_headers else None
+                )
+    
+    # Handle tags - can be either IDs (list of strings) or names (for lookup)
+    tag_ids_to_use = None
+    if tag_ids:
+        # Check if these are IDs (UUIDs) or names
+        if isinstance(tag_ids, list) and len(tag_ids) > 0:
+            # Assume IDs if first item looks like UUID (has hyphens and is long)
+            first_item = str(tag_ids[0])
+            if '-' in first_item and len(first_item) > 30:
+                tag_ids_to_use = tag_ids
+            elif lookup_tags_func:
+                # These are names, need to lookup/create
+                extra_headers = {}
+                if 'wix-site-id' in headers:
+                    extra_headers['wix-site-id'] = headers['wix-site-id']
+                tag_ids_to_use = lookup_tags_func(
+                    access_token, tag_ids, extra_headers if extra_headers else None
+                )
+    
+    # Add categories if we have IDs (must be non-empty list of strings)
+    # CRITICAL: Wix API rejects empty arrays or arrays with None/empty strings
+    if category_ids_to_use and isinstance(category_ids_to_use, list) and len(category_ids_to_use) > 0:
+        # Filter out None, empty strings, and ensure all are valid UUID strings
+        valid_category_ids = [str(cid).strip() for cid in category_ids_to_use if cid and str(cid).strip()]
+        if valid_category_ids:
+            blog_data['draftPost']['categoryIds'] = valid_category_ids
+            logger.debug(f"Added {len(valid_category_ids)} category IDs")
+        else:
+            logger.warning("All category IDs were invalid, not including categoryIds in payload")
+    
+    # Add tags if we have IDs (must be non-empty list of strings)
+    # CRITICAL: Wix API rejects empty arrays or arrays with None/empty strings
+    if tag_ids_to_use and isinstance(tag_ids_to_use, list) and len(tag_ids_to_use) > 0:
+        # Filter out None, empty strings, and ensure all are valid UUID strings
+        valid_tag_ids = [str(tid).strip() for tid in tag_ids_to_use if tid and str(tid).strip()]
+        if valid_tag_ids:
+            blog_data['draftPost']['tagIds'] = valid_tag_ids
+            logger.debug(f"Added {len(valid_tag_ids)} tag IDs")
+        else:
+            logger.warning("All tag IDs were invalid, not including tagIds in payload")
+    
+    # Build SEO data from metadata if provided
+    # TESTING: Skip SEO data temporarily to confirm richContent fix
+    test_skip_seo = True
+    if test_skip_seo:
+        logger.warning("🧪 TESTING: Skipping SEO data to isolate richContent vs seoData issue")
+        seo_data = None
+    elif seo_metadata:
+        logger.warning(f"📊 Building SEO data from metadata. Keys: {list(seo_metadata.keys())}")
+        seo_data = build_seo_data(seo_metadata, title)
+        if seo_data:
+            # Log detailed SEO structure
+            logger.warning(f"📋 SEO data built: {len(seo_data.get('tags', []))} tags, {len(seo_data.get('settings', {}).get('keywords', []))} keywords")
+            
+            # Log each SEO tag for debugging (key ones only to avoid too much output)
+            if seo_data.get('tags'):
+                for idx, tag in enumerate(seo_data['tags'][:3]):  # First 3 tags only
+                    tag_type = tag.get('type')
+                    if tag_type == 'title':
+                        logger.warning(f"  SEO tag {idx+1}: type={tag_type}, children={str(tag.get('children', ''))[:50]}...")
+                    else:
+                        props = tag.get('props', {})
+                        content_preview = str(props.get('content', props.get('href', props.get('name', ''))))[:50]
+                        logger.warning(f"  SEO tag {idx+1}: type={tag_type}, props={list(props.keys())}, content={content_preview}...")
+                if len(seo_data['tags']) > 3:
+                    logger.warning(f"  ... and {len(seo_data['tags']) - 3} more SEO tags")
+            
+            blog_data['draftPost']['seoData'] = seo_data
+            logger.warning(f"✅ Added seoData to blog post with {len(seo_data.get('tags', []))} tags")
+        else:
+            logger.warning("⚠️ SEO data was empty after building - check build_seo_data function")
+        
+        # Add SEO slug if provided (separate field from seoData)
+        if seo_metadata and seo_metadata.get('url_slug'):
+            blog_data['draftPost']['seoSlug'] = str(seo_metadata.get('url_slug')).strip()
+            logger.warning(f"✅ Added SEO slug: {blog_data['draftPost']['seoSlug']}")
+    
+    if test_skip_seo:
+        logger.warning("⚠️ SEO data skipped for testing - will add back once richContent is confirmed working")
+    elif not seo_metadata:
+        logger.warning("⚠️ No SEO metadata provided to create_blog_post")
+    
+    # Log the payload structure for debugging (without sensitive data)
+    logger.warning(f"📝 Creating blog post with title: '{title}'")
+    logger.warning(f"📋 Draft post fields: {list(blog_data['draftPost'].keys())}")
+    
+    # Detailed SEO logging
+    if 'seoData' in blog_data['draftPost']:
+        seo_data_debug = blog_data['draftPost']['seoData']
+        logger.warning(f"📊 SEO data in payload: {len(seo_data_debug.get('tags', []))} tags, {len(seo_data_debug.get('settings', {}).get('keywords', []))} keywords")
+        
+        # Log sample SEO tags (first 2 only to avoid too much output)
+        if seo_data_debug.get('tags'):
+            logger.warning("📋 SEO Tags sample:")
+            for i, tag in enumerate(seo_data_debug['tags'][:2]):  # First 2 tags
+                logger.warning(f"  Tag {i+1}: type={tag.get('type')}, custom={tag.get('custom')}, disabled={tag.get('disabled')}")
+            if len(seo_data_debug['tags']) > 2:
+                logger.warning(f"  ... and {len(seo_data_debug['tags']) - 2} more tags")
+        
+        if seo_data_debug.get('settings', {}).get('keywords'):
+            keywords_list = [k.get('term') for k in seo_data_debug['settings']['keywords'][:3]]
+            logger.warning(f"🔑 Keywords: {keywords_list}")
+        
+        # Log FULL seoData structure for debugging
+        import json
+        try:
+            seo_json = json.dumps(seo_data_debug, indent=2, ensure_ascii=False)
+            logger.warning(f"📄 FULL seoData JSON:\n{seo_json[:2000]}...")  # First 2000 chars
+        except Exception as e:
+            logger.error(f"Failed to serialize seoData: {e}")
+    else:
+        logger.warning("⚠️ No seoData in draft post payload!")
+    
+    try:
+        # Add wix-site-id header if we can extract it from token
+        extra_headers = {}
+        try:
+            token_str = str(access_token)
+            if token_str and token_str.startswith('OauthNG.JWS.'):
+                jwt_part = token_str[12:]
+                payload = jwt.decode(jwt_part, options={"verify_signature": False, "verify_aud": False})
+                data_payload = payload.get('data', {})
+                if isinstance(data_payload, str):
+                    try:
+                        data_payload = json.loads(data_payload)
+                    except:
+                        pass
+                instance_data = data_payload.get('instance', {})
+                meta_site_id = instance_data.get('metaSiteId')
+                if isinstance(meta_site_id, str) and meta_site_id:
+                    extra_headers['wix-site-id'] = meta_site_id
+                    headers['wix-site-id'] = meta_site_id
+        except Exception as e:
+            logger.debug(f"Could not extract site ID from token: {e}")
+        
+        # Make the API call
+        logger.warning(f"🚀 Calling Wix API: POST /blog/v3/draft-posts")
+        logger.warning(f"📦 Payload: title='{blog_data['draftPost'].get('title')}', has_seoData={'seoData' in blog_data['draftPost']}, has_richContent={'richContent' in blog_data['draftPost']}")
+        
+        # Validate payload structure before sending
+        draft_post = blog_data.get('draftPost', {})
+        if not isinstance(draft_post, dict):
+            raise ValueError("draftPost must be a dict object")
+        
+        # Validate richContent structure
+        if 'richContent' in draft_post:
+            rc = draft_post['richContent']
+            if not isinstance(rc, dict):
+                raise ValueError(f"richContent must be a dict, got {type(rc)}")
+            if 'nodes' not in rc:
+                raise ValueError("richContent missing 'nodes' field")
+            if not isinstance(rc['nodes'], list):
+                raise ValueError(f"richContent.nodes must be a list, got {type(rc['nodes'])}")
+            logger.debug(f"✅ richContent validation passed: {len(rc.get('nodes', []))} nodes")
+        
+        # Validate seoData structure if present
+        if 'seoData' in draft_post:
+            seo = draft_post['seoData']
+            if not isinstance(seo, dict):
+                raise ValueError(f"seoData must be a dict, got {type(seo)}")
+            if 'tags' in seo and not isinstance(seo['tags'], list):
+                raise ValueError(f"seoData.tags must be a list, got {type(seo.get('tags'))}")
+            if 'settings' in seo and not isinstance(seo['settings'], dict):
+                raise ValueError(f"seoData.settings must be a dict, got {type(seo.get('settings'))}")
+            logger.debug(f"✅ seoData validation passed: {len(seo.get('tags', []))} tags")
+        
+        # Final validation: Ensure no None values in any nested objects
+        # Wix API rejects None values and expects proper types
+        try:
+            validate_payload_no_none(blog_data, "blog_data")
+            logger.debug("✅ Payload validation passed: No None values found")
+        except ValueError as e:
+            logger.error(f"❌ Payload validation failed: {e}")
+            raise
+        
+        # Log full payload structure for debugging (sanitized)
+        logger.warning(f"📦 Full payload structure validation:")
+        logger.warning(f"  - draftPost type: {type(draft_post)}")
+        logger.warning(f"  - draftPost keys: {list(draft_post.keys())}")
+        logger.warning(f"  - richContent type: {type(draft_post.get('richContent'))}")
+        if 'richContent' in draft_post:
+            rc = draft_post['richContent']
+            logger.warning(f"  - richContent keys: {list(rc.keys()) if isinstance(rc, dict) else 'N/A'}")
+            logger.warning(f"  - richContent.nodes type: {type(rc.get('nodes'))}, count: {len(rc.get('nodes', []))}")
+            logger.warning(f"  - richContent.metadata type: {type(rc.get('metadata'))}")
+            logger.warning(f"  - richContent.documentStyle type: {type(rc.get('documentStyle'))}")
+        logger.warning(f"  - seoData type: {type(draft_post.get('seoData'))}")
+        if 'seoData' in draft_post:
+            seo = draft_post['seoData']
+            logger.warning(f"  - seoData keys: {list(seo.keys()) if isinstance(seo, dict) else 'N/A'}")
+            logger.warning(f"  - seoData.tags type: {type(seo.get('tags'))}, count: {len(seo.get('tags', []))}")
+            logger.warning(f"  - seoData.settings type: {type(seo.get('settings'))}")
+        if 'categoryIds' in draft_post:
+            logger.warning(f"  - categoryIds type: {type(draft_post.get('categoryIds'))}, count: {len(draft_post.get('categoryIds', []))}")
+        if 'tagIds' in draft_post:
+            logger.warning(f"  - tagIds type: {type(draft_post.get('tagIds'))}, count: {len(draft_post.get('tagIds', []))}")
+        
+        # Log a sample of the payload JSON to see exact structure (first 2000 chars)
+        try:
+            import json
+            payload_json = json.dumps(blog_data, indent=2, ensure_ascii=False)
+            logger.warning(f"📄 Payload JSON preview (first 3000 chars):\n{payload_json[:3000]}...")
+            
+            # Also log a deep structure inspection of richContent.nodes (first few nodes)
+            if 'richContent' in blog_data['draftPost']:
+                nodes = blog_data['draftPost']['richContent'].get('nodes', [])
+                if nodes:
+                    logger.warning(f"🔍 Inspecting first 5 richContent.nodes:")
+                    for i, node in enumerate(nodes[:5]):
+                        logger.warning(f"  Node {i+1}: type={node.get('type')}, keys={list(node.keys())}")
+                        # Check for any None values in node
+                        for key, value in node.items():
+                            if value is None:
+                                logger.error(f"    ⚠️ Node {i+1}.{key} is None!")
+                            elif isinstance(value, dict):
+                                for k, v in value.items():
+                                    if v is None:
+                                        logger.error(f"    ⚠️ Node {i+1}.{key}.{k} is None!")
+                        # Deep check: if it's a list-type node, inspect list items
+                        if node.get('type') in ['BULLETED_LIST', 'ORDERED_LIST']:
+                            list_items = node.get('nodes', [])
+                            if list_items:
+                                logger.warning(f"    List has {len(list_items)} items, checking first LIST_ITEM:")
+                                first_item = list_items[0]
+                                logger.warning(f"      LIST_ITEM keys: {list(first_item.keys())}")
+                                # Verify listItemData is NOT present (correct per Wix API spec)
+                                if 'listItemData' in first_item:
+                                    logger.error(f"      ❌ LIST_ITEM incorrectly has listItemData!")
+                                else:
+                                    logger.debug(f"      ✅ LIST_ITEM correctly has no listItemData")
+                                # Check nested PARAGRAPH nodes
+                                nested_nodes = first_item.get('nodes', [])
+                                if nested_nodes:
+                                    logger.warning(f"      LIST_ITEM has {len(nested_nodes)} nested nodes")
+                                    for n_idx, n_node in enumerate(nested_nodes[:2]):
+                                        logger.warning(f"        Nested node {n_idx+1}: type={n_node.get('type')}, keys={list(n_node.keys())}")
+        except Exception as e:
+            logger.warning(f"Could not serialize payload for logging: {e}")
+        
+        # Note: All node validation is done by validate_ricos_content() which runs earlier
+        # The recursive validation ensures all required data fields are present at any depth
+        
+        # Final deep validation: Serialize and deserialize to catch any JSON-serialization issues
+        # This will raise an error if there are any objects that can't be serialized
+        try:
+            import json
+            test_json = json.dumps(blog_data, ensure_ascii=False)
+            test_parsed = json.loads(test_json)
+            logger.debug("✅ Payload JSON serialization test passed")
+        except (TypeError, ValueError) as e:
+            logger.error(f"❌ Payload JSON serialization failed: {e}")
+            raise ValueError(f"Payload contains non-serializable data: {e}")
+        
+        # Final check: Ensure documentStyle and metadata are valid objects (not None, not empty strings)
+        rc = blog_data['draftPost']['richContent']
+        if 'documentStyle' in rc:
+            doc_style = rc['documentStyle']
+            if doc_style is None or doc_style == "":
+                logger.warning("⚠️ documentStyle is None or empty string, removing it")
+                del rc['documentStyle']
+            elif not isinstance(doc_style, dict):
+                logger.warning(f"⚠️ documentStyle is not a dict ({type(doc_style)}), removing it")
+                del rc['documentStyle']
+        
+        if 'metadata' in rc:
+            metadata = rc['metadata']
+            if metadata is None or metadata == "":
+                logger.warning("⚠️ metadata is None or empty string, removing it")
+                del rc['metadata']
+            elif not isinstance(metadata, dict):
+                logger.warning(f"⚠️ metadata is not a dict ({type(metadata)}), removing it")
+                del rc['metadata']
+        
+        # Check for any None values in critical nested structures
+        def check_none_in_dict(d, path=""):
+            """Recursively check for None values that shouldn't be there"""
+            issues = []
+            if isinstance(d, dict):
+                for key, value in d.items():
+                    current_path = f"{path}.{key}" if path else key
+                    if value is None:
+                        # Some fields can legitimately be None, but most shouldn't
+                        if key not in ['decorations', 'nodeStyle', 'props']:
+                            issues.append(current_path)
+                    elif isinstance(value, dict):
+                        issues.extend(check_none_in_dict(value, current_path))
+                    elif isinstance(value, list):
+                        for i, item in enumerate(value):
+                            if item is None:
+                                issues.append(f"{current_path}[{i}]")
+                            elif isinstance(item, dict):
+                                issues.extend(check_none_in_dict(item, f"{current_path}[{i}]"))
+            return issues
+        
+        none_issues = check_none_in_dict(blog_data['draftPost']['richContent'])
+        if none_issues:
+            logger.error(f"❌ Found None values in richContent at: {none_issues[:10]}")  # Limit to first 10
+            # Remove None values from critical paths
+            for issue_path in none_issues[:5]:  # Fix first 5
+                parts = issue_path.split('.')
+                try:
+                    obj = blog_data['draftPost']['richContent']
+                    for part in parts[:-1]:
+                        if '[' in part:
+                            key, idx = part.split('[')
+                            idx = int(idx.rstrip(']'))
+                            obj = obj[key][idx]
+                        else:
+                            obj = obj[part]
+                    final_key = parts[-1]
+                    if '[' in final_key:
+                        key, idx = final_key.split('[')
+                        idx = int(idx.rstrip(']'))
+                        obj[key][idx] = {}
+                    else:
+                        obj[final_key] = {}
+                    logger.warning(f"Fixed None value at {issue_path}")
+                except:
+                    pass
+        
+        # Log the final payload structure one more time before sending
+        logger.warning(f"📤 Final payload ready - draftPost keys: {list(blog_data['draftPost'].keys())}")
+        logger.warning(f"📤 RichContent nodes count: {len(blog_data['draftPost']['richContent'].get('nodes', []))}")
+        logger.warning(f"📤 RichContent has metadata: {bool(blog_data['draftPost']['richContent'].get('metadata'))}")
+        logger.warning(f"📤 RichContent has documentStyle: {bool(blog_data['draftPost']['richContent'].get('documentStyle'))}")
+        
+        # Try sending WITHOUT SEO data first to isolate the issue
+        test_without_seo = False  # Disabled - listItemData issue fixed
+        if test_without_seo and 'seoData' in blog_data['draftPost']:
+            logger.warning("🧪 TESTING WITHOUT SEO DATA to isolate issue...")
+            # Clone the payload without SEO data
+            test_payload_no_seo = {
+                'draftPost': {
+                    'title': blog_data['draftPost']['title'],
+                    'memberId': blog_data['draftPost']['memberId'],
+                    'richContent': blog_data['draftPost']['richContent'],
+                    'excerpt': blog_data['draftPost'].get('excerpt', '')
+                },
+                'publish': False,
+                'fieldsets': ['URL']
+            }
+            try:
+                logger.warning("🧪 Attempting without SEO data...")
+                test_result = blog_service.create_draft_post(access_token, test_payload_no_seo, extra_headers or None)
+                logger.warning(f"✅ WITHOUT SEO DATA SUCCEEDED! Post ID: {test_result.get('draftPost', {}).get('id')}")
+                logger.error("⚠️⚠️⚠️ ISSUE IS WITH SEO DATA STRUCTURE!")
+                # If this succeeds, don't send the full payload, just return this result
+                return test_result
+            except Exception as e:
+                logger.warning(f"❌ WITHOUT SEO DATA ALSO FAILED: {e}")
+                logger.warning("⚠️ Issue is NOT with SEO data, continuing with full payload...")
+        
+        # Try sending with minimal structure first to isolate the issue
+        # Create a test payload with just required fields
+        minimal_test = False  # Set to True to test with minimal payload
+        if minimal_test:
+            logger.warning("🧪 TESTING WITH MINIMAL PAYLOAD (title + memberId + simple richContent)")
+            test_payload = {
+                'draftPost': {
+                    'title': blog_data['draftPost']['title'],
+                    'memberId': blog_data['draftPost']['memberId'],
+                    'richContent': {
+                        'nodes': [
+                            {
+                                'id': str(uuid.uuid4()),
+                                'type': 'PARAGRAPH',
+                                'nodes': [
+                                    {
+                                        'id': str(uuid.uuid4()),
+                                        'type': 'TEXT',
+                                        'textData': {
+                                            'text': 'Test paragraph',
+                                            'decorations': []
+                                        }
+                                    }
+                                ],
+                                'paragraphData': {}
+                            }
+                        ],
+                        'metadata': {'version': 1, 'id': str(uuid.uuid4())},
+                        'documentStyle': {}
+                    }
+                },
+                'publish': False,
+                'fieldsets': ['URL']
+            }
+            logger.warning("🧪 Attempting minimal payload first...")
+            try:
+                test_result = blog_service.create_draft_post(access_token, test_payload, extra_headers or None)
+                logger.warning(f"✅ MINIMAL PAYLOAD SUCCEEDED! Post ID: {test_result.get('draftPost', {}).get('id')}")
+                logger.warning("⚠️ Issue is with complex content, not basic structure")
+            except Exception as e:
+                logger.error(f"❌ MINIMAL PAYLOAD ALSO FAILED: {e}")
+                logger.error("⚠️ Issue is with basic structure or permissions")
+        
+        result = blog_service.create_draft_post(access_token, blog_data, extra_headers or None)
+        
+        # Log response
+        draft_post = result.get('draftPost', {})
+        logger.warning(f"✅ Blog post created successfully! Post ID: {draft_post.get('id', 'N/A')}")
+        
+        # Check if SEO data was preserved in response
+        if 'seoData' in draft_post:
+            seo_response = draft_post['seoData']
+            logger.warning(f"✅ SEO data confirmed in response: {len(seo_response.get('tags', []))} tags, {len(seo_response.get('settings', {}).get('keywords', []))} keywords")
+        else:
+            logger.warning("⚠️ No seoData in response - it may have been filtered out by Wix API")
+            logger.warning(f"📋 Response fields: {list(draft_post.keys())}")
+        
+        return result
+    except requests.RequestException as e:
+        logger.error(f"Failed to create blog post: {e}")
+        if hasattr(e, 'response') and e.response is not None:
+            logger.error(f"Response body: {e.response.text}")
+        raise
+
diff --git a/backend/services/integrations/wix/content.py b/backend/services/integrations/wix/content.py
index 216b19df..0a31aec4 100644
--- a/backend/services/integrations/wix/content.py
+++ b/backend/services/integrations/wix/content.py
@@ -1,58 +1,460 @@
+import re
+import uuid
 from typing import Any, Dict, List
 
 
+def parse_markdown_inline(text: str) -> List[Dict[str, Any]]:
+    """
+    Parse inline markdown formatting (bold, italic, links) into Ricos text nodes.
+    Returns a list of text nodes with decorations.
+    Handles: **bold**, *italic*, [links](url), `code`, and combinations.
+    """
+    if not text:
+        return [{
+            'id': str(uuid.uuid4()),
+            'type': 'TEXT',
+            'textData': {'text': '', 'decorations': []}
+        }]
+    
+    nodes = []
+    
+    # Process text character by character to handle nested/adjacent formatting
+    # This is more robust than regex for complex cases
+    i = 0
+    current_text = ''
+    current_decorations = []
+    
+    while i < len(text):
+        # Check for bold **text** (must come before single * check)
+        if i < len(text) - 1 and text[i:i+2] == '**':
+            # Save any accumulated text
+            if current_text:
+                nodes.append({
+                    'id': str(uuid.uuid4()),
+                    'type': 'TEXT',
+                    'textData': {
+                        'text': current_text,
+                        'decorations': current_decorations.copy()
+                    }
+                })
+                current_text = ''
+            
+            # Find closing **
+            end_bold = text.find('**', i + 2)
+            if end_bold != -1:
+                bold_text = text[i + 2:end_bold]
+                # Recursively parse the bold text for nested formatting
+                bold_nodes = parse_markdown_inline(bold_text)
+                # Add BOLD decoration to all text nodes within
+                for node in bold_nodes:
+                    if node['type'] == 'TEXT':
+                        node_decorations = node['textData'].get('decorations', []).copy()
+                        if 'BOLD' not in node_decorations:
+                            node_decorations.append('BOLD')
+                        node['textData']['decorations'] = node_decorations
+                    nodes.append(node)
+                i = end_bold + 2
+                continue
+        
+        # Check for link [text](url)
+        elif text[i] == '[':
+            # Save any accumulated text
+            if current_text:
+                nodes.append({
+                    'id': str(uuid.uuid4()),
+                    'type': 'TEXT',
+                    'textData': {
+                        'text': current_text,
+                        'decorations': current_decorations.copy()
+                    }
+                })
+                current_text = ''
+                current_decorations = []
+            
+            # Find matching ]
+            link_end = text.find(']', i)
+            if link_end != -1 and link_end < len(text) - 1 and text[link_end + 1] == '(':
+                link_text = text[i + 1:link_end]
+                url_start = link_end + 2
+                url_end = text.find(')', url_start)
+                if url_end != -1:
+                    url = text[url_start:url_end]
+                    # Create link node
+                    link_node_id = str(uuid.uuid4())
+                    text_node_id = str(uuid.uuid4())
+                    link_text_nodes = parse_markdown_inline(link_text)
+                    # Wrap link text in LINK node
+                    nodes.append({
+                        'id': link_node_id,
+                        'type': 'LINK',
+                        'nodes': link_text_nodes if link_text_nodes else [{
+                            'id': text_node_id,
+                            'type': 'TEXT',
+                            'textData': {'text': link_text, 'decorations': []}
+                        }],
+                        'linkData': {
+                            'link': {
+                                'url': url,
+                                'target': '_blank'
+                            }
+                        }
+                    })
+                    i = url_end + 1
+                    continue
+        
+        # Check for code `text`
+        elif text[i] == '`':
+            # Save any accumulated text
+            if current_text:
+                nodes.append({
+                    'id': str(uuid.uuid4()),
+                    'type': 'TEXT',
+                    'textData': {
+                        'text': current_text,
+                        'decorations': current_decorations.copy()
+                    }
+                })
+                current_text = ''
+                current_decorations = []
+            
+            # Find closing `
+            code_end = text.find('`', i + 1)
+            if code_end != -1:
+                code_text = text[i + 1:code_end]
+                nodes.append({
+                    'id': str(uuid.uuid4()),
+                    'type': 'TEXT',
+                    'textData': {
+                        'text': code_text,
+                        'decorations': ['CODE']
+                    }
+                })
+                i = code_end + 1
+                continue
+        
+        # Check for italic *text* (only if not part of **)
+        elif text[i] == '*' and (i == 0 or text[i-1] != '*') and (i == len(text) - 1 or text[i+1] != '*'):
+            # Save any accumulated text
+            if current_text:
+                nodes.append({
+                    'id': str(uuid.uuid4()),
+                    'type': 'TEXT',
+                    'textData': {
+                        'text': current_text,
+                        'decorations': current_decorations.copy()
+                    }
+                })
+                current_text = ''
+                current_decorations = []
+            
+            # Find closing * (but not **)
+            italic_end = text.find('*', i + 1)
+            if italic_end != -1:
+                # Make sure it's not part of **
+                if italic_end == len(text) - 1 or text[italic_end + 1] != '*':
+                    italic_text = text[i + 1:italic_end]
+                    italic_nodes = parse_markdown_inline(italic_text)
+                    # Add ITALIC decoration
+                    for node in italic_nodes:
+                        if node['type'] == 'TEXT':
+                            node_decorations = node['textData'].get('decorations', []).copy()
+                            if 'ITALIC' not in node_decorations:
+                                node_decorations.append('ITALIC')
+                            node['textData']['decorations'] = node_decorations
+                        nodes.append(node)
+                    i = italic_end + 1
+                    continue
+        
+        # Regular character
+        current_text += text[i]
+        i += 1
+    
+    # Add any remaining text
+    if current_text:
+        nodes.append({
+            'id': str(uuid.uuid4()),
+            'type': 'TEXT',
+            'textData': {
+                'text': current_text,
+                'decorations': current_decorations.copy()
+            }
+        })
+    
+    # If no nodes created, return single plain text node
+    if not nodes:
+        nodes.append({
+            'id': str(uuid.uuid4()),
+            'type': 'TEXT',
+            'textData': {
+                'text': text,
+                'decorations': []
+            }
+        })
+    
+    return nodes
+
+
 def convert_content_to_ricos(content: str, images: List[str] = None) -> Dict[str, Any]:
     """
-    Convert simple markdown-like text into minimal valid Ricos JSON.
+    Convert markdown content into valid Ricos JSON format.
+    Supports headings, paragraphs, lists, bold, italic, links, and images.
     """
-    paragraphs = content.split('\n\n')
+    if not content:
+        content = "This is a post from ALwrity."
+    
     nodes = []
-
-    import uuid
-
-    for paragraph in paragraphs:
-        text = paragraph.strip()
-        if not text:
+    lines = content.split('\n')
+    
+    i = 0
+    while i < len(lines):
+        line = lines[i].strip()
+        
+        if not line:
+            i += 1
             continue
+        
         node_id = str(uuid.uuid4())
-        text_node_id = str(uuid.uuid4())
-
-        if text.startswith('#'):
-            level = len(text) - len(text.lstrip('#'))
-            heading_text = text.lstrip('# ').strip()
+        
+        # Check for headings
+        if line.startswith('#'):
+            level = len(line) - len(line.lstrip('#'))
+            heading_text = line.lstrip('# ').strip()
+            text_nodes = parse_markdown_inline(heading_text)
             nodes.append({
                 'id': node_id,
                 'type': 'HEADING',
-                'nodes': [{
-                    'id': text_node_id,
-                    'type': 'TEXT',
-                    'textData': {
-                        'text': heading_text,
-                        'decorations': []
-                    }
-                }],
-                'headingData': { 'level': min(level, 6) }
+                'nodes': text_nodes,
+                'headingData': {'level': min(level, 6)}
             })
-        else:
-            nodes.append({
-                'id': node_id,
+            i += 1
+        
+        # Check for blockquotes
+        elif line.startswith('>'):
+            quote_text = line.lstrip('> ').strip()
+            # Continue reading consecutive blockquote lines
+            quote_lines = [quote_text]
+            i += 1
+            while i < len(lines) and lines[i].strip().startswith('>'):
+                quote_lines.append(lines[i].strip().lstrip('> ').strip())
+                i += 1
+            quote_content = ' '.join(quote_lines)
+            text_nodes = parse_markdown_inline(quote_content)
+            # CRITICAL: TEXT nodes must be wrapped in PARAGRAPH nodes within BLOCKQUOTE
+            paragraph_node = {
+                'id': str(uuid.uuid4()),
                 'type': 'PARAGRAPH',
-                'nodes': [{
-                    'id': text_node_id,
-                    'type': 'TEXT',
-                    'textData': {
-                        'text': text,
-                        'decorations': []
-                    }
-                }],
+                'nodes': text_nodes,
                 'paragraphData': {}
-            })
-
+            }
+            blockquote_node = {
+                'id': node_id,
+                'type': 'BLOCKQUOTE',
+                'nodes': [paragraph_node],
+                'blockquoteData': {}
+            }
+            nodes.append(blockquote_node)
+        
+        # Check for unordered lists (handle both '- ' and '* ' markers)
+        elif (line.startswith('- ') or line.startswith('* ') or 
+             (line.startswith('-') and len(line) > 1 and line[1] != '-') or
+             (line.startswith('*') and len(line) > 1 and line[1] != '*')):
+            list_items = []
+            list_marker = '- ' if line.startswith('-') else '* '
+            # Process list items
+            while i < len(lines):
+                current_line = lines[i].strip()
+                # Check if this is a list item
+                is_list_item = (current_line.startswith('- ') or current_line.startswith('* ') or
+                               (current_line.startswith('-') and len(current_line) > 1 and current_line[1] != '-') or
+                               (current_line.startswith('*') and len(current_line) > 1 and current_line[1] != '*'))
+                
+                if not is_list_item:
+                    break
+                
+                # Extract item text (handle both '- ' and '-item' formats)
+                if current_line.startswith('- ') or current_line.startswith('* '):
+                    item_text = current_line[2:].strip()
+                elif current_line.startswith('-'):
+                    item_text = current_line[1:].strip()
+                elif current_line.startswith('*'):
+                    item_text = current_line[1:].strip()
+                else:
+                    item_text = current_line
+                
+                list_items.append(item_text)
+                i += 1
+                
+                # Check for nested items (indented with 2+ spaces)
+                while i < len(lines):
+                    next_line = lines[i]
+                    # Must be indented and be a list marker
+                    if next_line.startswith('  ') and (next_line.strip().startswith('- ') or 
+                                                      next_line.strip().startswith('* ') or
+                                                      (next_line.strip().startswith('-') and len(next_line.strip()) > 1) or
+                                                      (next_line.strip().startswith('*') and len(next_line.strip()) > 1)):
+                        nested_text = next_line.strip()
+                        if nested_text.startswith('- ') or nested_text.startswith('* '):
+                            nested_text = nested_text[2:].strip()
+                        elif nested_text.startswith('-'):
+                            nested_text = nested_text[1:].strip()
+                        elif nested_text.startswith('*'):
+                            nested_text = nested_text[1:].strip()
+                        list_items.append(nested_text)
+                        i += 1
+                    else:
+                        break
+            
+            # Build list items with proper formatting
+            # CRITICAL: TEXT nodes must be wrapped in PARAGRAPH nodes within LIST_ITEM
+            # NOTE: LIST_ITEM nodes do NOT have a data field per Wix API schema
+            # Wix API: omit empty data objects, don't include them as {}
+            list_node_items = []
+            for item_text in list_items:
+                item_node_id = str(uuid.uuid4())
+                text_nodes = parse_markdown_inline(item_text)
+                paragraph_node = {
+                    'id': str(uuid.uuid4()),
+                    'type': 'PARAGRAPH',
+                    'nodes': text_nodes,
+                    'paragraphData': {}
+                }
+                list_item_node = {
+                    'id': item_node_id,
+                    'type': 'LIST_ITEM',
+                    'nodes': [paragraph_node]
+                }
+                list_node_items.append(list_item_node)
+            
+            bulleted_list_node = {
+                'id': node_id,
+                'type': 'BULLETED_LIST',
+                'nodes': list_node_items,
+                'bulletedListData': {}
+            }
+            nodes.append(bulleted_list_node)
+        
+        # Check for ordered lists
+        elif re.match(r'^\d+\.\s+', line):
+            list_items = []
+            while i < len(lines) and re.match(r'^\d+\.\s+', lines[i].strip()):
+                item_text = re.sub(r'^\d+\.\s+', '', lines[i].strip())
+                list_items.append(item_text)
+                i += 1
+                # Check for nested items
+                while i < len(lines) and lines[i].strip().startswith('  ') and re.match(r'^\s+\d+\.\s+', lines[i].strip()):
+                    nested_text = re.sub(r'^\s+\d+\.\s+', '', lines[i].strip())
+                    list_items.append(nested_text)
+                    i += 1
+            
+            # CRITICAL: TEXT nodes must be wrapped in PARAGRAPH nodes within LIST_ITEM
+            # NOTE: LIST_ITEM nodes do NOT have a data field per Wix API schema
+            # Wix API: omit empty data objects, don't include them as {}
+            list_node_items = []
+            for item_text in list_items:
+                item_node_id = str(uuid.uuid4())
+                text_nodes = parse_markdown_inline(item_text)
+                paragraph_node = {
+                    'id': str(uuid.uuid4()),
+                    'type': 'PARAGRAPH',
+                    'nodes': text_nodes,
+                    'paragraphData': {}
+                }
+                list_item_node = {
+                    'id': item_node_id,
+                    'type': 'LIST_ITEM',
+                    'nodes': [paragraph_node]
+                }
+                list_node_items.append(list_item_node)
+            
+            ordered_list_node = {
+                'id': node_id,
+                'type': 'ORDERED_LIST',
+                'nodes': list_node_items,
+                'orderedListData': {}
+            }
+            nodes.append(ordered_list_node)
+        
+        # Check for images
+        elif line.startswith('!['):
+            img_match = re.match(r'!\[([^\]]*)\]\(([^)]+)\)', line)
+            if img_match:
+                alt_text = img_match.group(1)
+                img_url = img_match.group(2)
+                nodes.append({
+                    'id': node_id,
+                    'type': 'IMAGE',
+                    'nodes': [],
+                    'imageData': {
+                        'image': {
+                            'src': {'url': img_url},
+                            'altText': alt_text
+                        },
+                        'containerData': {
+                            'alignment': 'CENTER',
+                            'width': {'size': 'CONTENT'}
+                        }
+                    }
+                })
+            i += 1
+        
+        # Regular paragraph
+        else:
+            # Collect consecutive non-empty lines as paragraph content
+            para_lines = [line]
+            i += 1
+            while i < len(lines):
+                next_line = lines[i].strip()
+                if not next_line:
+                    break
+                # Stop if next line is a special markdown element
+                if (next_line.startswith('#') or 
+                    next_line.startswith('- ') or 
+                    next_line.startswith('* ') or
+                    next_line.startswith('>') or
+                    next_line.startswith('![') or
+                    re.match(r'^\d+\.\s+', next_line)):
+                    break
+                para_lines.append(next_line)
+                i += 1
+            
+            para_text = ' '.join(para_lines)
+            text_nodes = parse_markdown_inline(para_text)
+            
+            # Only add paragraph if there are text nodes
+            if text_nodes:
+                paragraph_node = {
+                    'id': node_id,
+                    'type': 'PARAGRAPH',
+                    'nodes': text_nodes,
+                    'paragraphData': {}
+                }
+                nodes.append(paragraph_node)
+    
+    # Ensure at least one node exists
+    # Wix API: omit empty data objects, don't include them as {}
+    if not nodes:
+        fallback_paragraph = {
+            'id': str(uuid.uuid4()),
+            'type': 'PARAGRAPH',
+            'nodes': [{
+                'id': str(uuid.uuid4()),
+                'type': 'TEXT',
+                'textData': {
+                    'text': content[:500] if content else "This is a post from ALwrity.",
+                    'decorations': []
+                }
+            }],
+            'paragraphData': {}
+        }
+        nodes.append(fallback_paragraph)
+    
     return {
+        'type': 'DOCUMENT',
+        'id': str(uuid.uuid4()),
         'nodes': nodes,
-        'metadata': { 'version': 1, 'id': str(uuid.uuid4()) },
+        'metadata': {'version': 1, 'id': str(uuid.uuid4())},
         'documentStyle': {
-            'paragraph': { 'decorations': [], 'nodeStyle': {}, 'lineHeight': '1.5' }
+            'paragraph': {'decorations': [], 'nodeStyle': {}, 'lineHeight': '1.5'}
         }
     }
 
diff --git a/backend/services/integrations/wix/media.py b/backend/services/integrations/wix/media.py
index afd1c2ef..eab4ba8b 100644
--- a/backend/services/integrations/wix/media.py
+++ b/backend/services/integrations/wix/media.py
@@ -7,6 +7,12 @@ class WixMediaService:
         self.base_url = base_url
 
     def import_image(self, access_token: str, image_url: str, display_name: str) -> Dict[str, Any]:
+        """
+        Import external image to Wix Media Manager.
+        
+        Official endpoint: https://www.wixapis.com/site-media/v1/files/import
+        Reference: https://dev.wix.com/docs/rest/assets/media/media-manager/files/import-file
+        """
         headers = {
             'Authorization': f'Bearer {access_token}',
             'Content-Type': 'application/json',
@@ -16,7 +22,9 @@ class WixMediaService:
             'mediaType': 'IMAGE',
             'displayName': display_name,
         }
-        response = requests.post(f"{self.base_url}/media/v1/files/import", headers=headers, json=payload)
+        # Correct endpoint per Wix API documentation
+        endpoint = f"{self.base_url}/site-media/v1/files/import"
+        response = requests.post(endpoint, headers=headers, json=payload)
         response.raise_for_status()
         return response.json()
 
diff --git a/backend/services/integrations/wix/ricos_converter.py b/backend/services/integrations/wix/ricos_converter.py
new file mode 100644
index 00000000..faba70c7
--- /dev/null
+++ b/backend/services/integrations/wix/ricos_converter.py
@@ -0,0 +1,277 @@
+"""
+Ricos Document Converter for Wix
+
+Converts markdown content to Wix Ricos JSON format using either:
+1. Wix's official Ricos Documents API (preferred)
+2. Custom markdown parser (fallback)
+"""
+
+import json
+import requests
+import jwt
+from typing import Dict, Any, Optional
+from loguru import logger
+
+
+def markdown_to_html(markdown_content: str) -> str:
+    """
+    Convert markdown content to HTML.
+    Uses a simple markdown parser for basic conversion.
+    
+    Args:
+        markdown_content: Markdown content to convert
+        
+    Returns:
+        HTML string
+    """
+    try:
+        # Try using markdown library if available
+        import markdown
+        html = markdown.markdown(markdown_content, extensions=['fenced_code', 'tables'])
+        return html
+    except ImportError:
+        # Fallback: Simple regex-based conversion for basic markdown
+        logger.warning("markdown library not available, using basic markdown-to-HTML conversion")
+        import re
+        
+        if not markdown_content or not markdown_content.strip():
+            return "<p>This is a post from ALwrity.</p>"
+        
+        lines = markdown_content.split('\n')
+        result = []
+        in_list = False
+        list_type = None  # 'ul' or 'ol'
+        in_code_block = False
+        code_block_content = []
+        
+        i = 0
+        while i < len(lines):
+            line = lines[i].strip()
+            
+            # Handle code blocks first
+            if line.startswith('```'):
+                if not in_code_block:
+                    in_code_block = True
+                    code_block_content = []
+                    i += 1
+                    continue
+                else:
+                    in_code_block = False
+                    result.append(f'<pre><code>{"\n".join(code_block_content)}</code></pre>')
+                    code_block_content = []
+                    i += 1
+                    continue
+            
+            if in_code_block:
+                code_block_content.append(lines[i])
+                i += 1
+                continue
+            
+            # Close any open lists
+            if in_list and not (line.startswith('- ') or line.startswith('* ') or re.match(r'^\d+\.\s+', line)):
+                result.append(f'</{list_type}>')
+                in_list = False
+                list_type = None
+            
+            if not line:
+                i += 1
+                continue
+            
+            # Headers
+            if line.startswith('###'):
+                result.append(f'<h3>{line[3:].strip()}</h3>')
+            elif line.startswith('##'):
+                result.append(f'<h2>{line[2:].strip()}</h2>')
+            elif line.startswith('#'):
+                result.append(f'<h1>{line[1:].strip()}</h1>')
+            # Lists
+            elif line.startswith('- ') or line.startswith('* '):
+                if not in_list or list_type != 'ul':
+                    if in_list:
+                        result.append(f'</{list_type}>')
+                    result.append('<ul>')
+                    in_list = True
+                    list_type = 'ul'
+                # Process inline formatting in list item
+                item_text = line[2:].strip()
+                item_text = re.sub(r'\*\*(.*?)\*\*', r'<strong>\1</strong>', item_text)
+                item_text = re.sub(r'\*(.*?)\*', r'<em>\1</em>', item_text)
+                result.append(f'<li>{item_text}</li>')
+            elif re.match(r'^\d+\.\s+', line):
+                if not in_list or list_type != 'ol':
+                    if in_list:
+                        result.append(f'</{list_type}>')
+                    result.append('<ol>')
+                    in_list = True
+                    list_type = 'ol'
+                # Process inline formatting in list item
+                match = re.match(r'^\d+\.\s+(.*)', line)
+                if match:
+                    item_text = match.group(1)
+                    item_text = re.sub(r'\*\*(.*?)\*\*', r'<strong>\1</strong>', item_text)
+                    item_text = re.sub(r'\*(.*?)\*', r'<em>\1</em>', item_text)
+                    result.append(f'<li>{item_text}</li>')
+            # Blockquotes
+            elif line.startswith('>'):
+                quote_text = line[1:].strip()
+                quote_text = re.sub(r'\*\*(.*?)\*\*', r'<strong>\1</strong>', quote_text)
+                quote_text = re.sub(r'\*(.*?)\*', r'<em>\1</em>', quote_text)
+                result.append(f'<blockquote><p>{quote_text}</p></blockquote>')
+            # Regular paragraphs
+            else:
+                para_text = line
+                # Process inline formatting
+                para_text = re.sub(r'\*\*(.*?)\*\*', r'<strong>\1</strong>', para_text)
+                para_text = re.sub(r'\*(.*?)\*', r'<em>\1</em>', para_text)
+                para_text = re.sub(r'\[([^\]]+)\]\(([^\)]+)\)', r'<a href="\2">\1</a>', para_text)
+                para_text = re.sub(r'`([^`]+)`', r'<code>\1</code>', para_text)
+                result.append(f'<p>{para_text}</p>')
+            
+            i += 1
+        
+        # Close any open lists
+        if in_list:
+            result.append(f'</{list_type}>')
+        
+        # Ensure we have at least one paragraph
+        if not result:
+            result.append('<p>This is a post from ALwrity.</p>')
+        
+        html = '\n'.join(result)
+        
+        logger.debug(f"Converted {len(markdown_content)} chars markdown to {len(html)} chars HTML")
+        return html
+
+
+def convert_via_wix_api(markdown_content: str, access_token: str, base_url: str = 'https://www.wixapis.com') -> Dict[str, Any]:
+    """
+    Convert markdown to Ricos using Wix's official Ricos Documents API.
+    Uses HTML format for better reliability (per Wix documentation, HTML is fully supported).
+    
+    Reference: https://dev.wix.com/docs/api-reference/assets/rich-content/ricos-documents/convert-to-ricos-document
+    
+    Args:
+        markdown_content: Markdown content to convert (will be converted to HTML)
+        access_token: Wix access token
+        base_url: Wix API base URL (default: https://www.wixapis.com)
+        
+    Returns:
+        Ricos JSON document
+    """
+    # Validate content is not empty
+    markdown_stripped = markdown_content.strip() if markdown_content else ""
+    if not markdown_stripped:
+        logger.error("Markdown content is empty or whitespace-only")
+        raise ValueError("Content cannot be empty for Wix Ricos API conversion")
+    
+    logger.debug(f"Converting markdown to HTML: input_length={len(markdown_stripped)} chars")
+    
+    # Convert markdown to HTML for better reliability with Wix API
+    # HTML format is more structured and less prone to parsing errors
+    html_content = markdown_to_html(markdown_stripped)
+    
+    # Validate HTML content is not empty - CRITICAL for Wix API
+    html_stripped = html_content.strip() if html_content else ""
+    if not html_stripped or len(html_stripped) == 0:
+        logger.error(f"HTML conversion produced empty content! Markdown length: {len(markdown_stripped)}")
+        logger.error(f"Markdown sample: {markdown_stripped[:500]}...")
+        logger.error(f"HTML result: '{html_content}' (type: {type(html_content)})")
+        # Fallback: use a minimal valid HTML if conversion failed
+        html_content = "<p>Content from ALwrity blog writer.</p>"
+        logger.warning("Using fallback HTML due to empty conversion result")
+    else:
+        html_content = html_stripped
+    
+    logger.debug(f"✅ Converted markdown to HTML: {len(html_content)} chars, preview: {html_content[:200]}...")
+    
+    headers = {
+        'Authorization': f'Bearer {access_token}',
+        'Content-Type': 'application/json'
+    }
+    
+    # Add wix-site-id if available from token
+    try:
+        token_str = str(access_token)
+        if token_str and token_str.startswith('OauthNG.JWS.'):
+            jwt_part = token_str[12:]
+            payload = jwt.decode(jwt_part, options={"verify_signature": False, "verify_aud": False})
+            data_payload = payload.get('data', {})
+            if isinstance(data_payload, str):
+                try:
+                    data_payload = json.loads(data_payload)
+                except:
+                    pass
+            instance_data = data_payload.get('instance', {})
+            meta_site_id = instance_data.get('metaSiteId')
+            if isinstance(meta_site_id, str) and meta_site_id:
+                headers['wix-site-id'] = meta_site_id
+    except Exception as e:
+        logger.debug(f"Could not extract site ID from token: {e}")
+    
+    # Call Wix Ricos Documents API: Convert to Ricos Document
+    # Official endpoint: https://www.wixapis.com/ricos/v1/ricos-document/convert/to-ricos
+    # Reference: https://dev.wix.com/docs/rest/assets/rich-content/ricos-documents/convert-to-ricos-document
+    endpoint = f"{base_url}/ricos/v1/ricos-document/convert/to-ricos"
+    
+    # Ensure HTML content is not empty or just whitespace
+    html_stripped = html_content.strip() if html_content else ""
+    if not html_stripped or len(html_stripped) == 0:
+        logger.error(f"HTML content is empty after conversion. Markdown length: {len(markdown_content)}")
+        logger.error(f"Markdown preview (first 500 chars): {markdown_content[:500] if markdown_content else 'N/A'}")
+        raise ValueError(f"HTML content cannot be empty. Original markdown had {len(markdown_content)} characters.")
+    
+    # Payload structure per Wix API: html/markdown/plainText field at root, optional plugins
+    payload = {
+        'html': html_stripped,  # Direct field, not nested in options
+        'plugins': []  # Optional: empty array uses default plugins
+    }
+    
+    logger.warning(f"📤 Sending to Wix Ricos API: html_length={len(payload['html'])}, plugins_count={len(payload['plugins'])}")
+    logger.debug(f"HTML preview (first 300 chars): {html_stripped[:300]}...")
+    
+    try:
+        # Log the exact payload being sent (for debugging)
+        logger.warning(f"📤 Wix Ricos API Request:")
+        logger.warning(f"  Endpoint: {endpoint}")
+        logger.warning(f"  Payload keys: {list(payload.keys())}")
+        logger.warning(f"  HTML length: {len(payload.get('html', ''))}")
+        logger.warning(f"  Plugins: {payload.get('plugins', [])}")
+        logger.debug(f"  Full payload (first 500 chars of HTML): {str(payload)[:500]}")
+        
+        response = requests.post(
+            endpoint,
+            headers=headers,
+            json=payload,
+            timeout=30
+        )
+        response.raise_for_status()
+        result = response.json()
+        
+        # Extract the ricos document from response
+        # Response structure: { "document": { "nodes": [...], "metadata": {...}, "documentStyle": {...} } }
+        ricos_document = result.get('document')
+        if not ricos_document:
+            # Fallback: try other possible response fields
+            ricos_document = result.get('ricosDocument') or result.get('ricos') or result
+        
+        if not ricos_document:
+            logger.error(f"Unexpected response structure from Wix API: {list(result.keys())}")
+            logger.error(f"Response: {result}")
+            raise ValueError("Wix API did not return a valid Ricos document")
+        
+        logger.warning(f"✅ Successfully converted HTML to Ricos via Wix API: {len(ricos_document.get('nodes', []))} nodes")
+        return ricos_document
+        
+    except requests.RequestException as e:
+        logger.error(f"❌ Wix Ricos API conversion failed: {e}")
+        if hasattr(e, 'response') and e.response is not None:
+            logger.error(f"  Response status: {e.response.status_code}")
+            logger.error(f"  Response headers: {dict(e.response.headers)}")
+            try:
+                error_body = e.response.json()
+                logger.error(f"  Response JSON: {error_body}")
+            except:
+                logger.error(f"  Response text: {e.response.text}")
+            logger.error(f"  Request payload was: {json.dumps(payload, indent=2)[:1000]}...")  # First 1000 chars
+        raise
+
diff --git a/backend/services/integrations/wix/seo.py b/backend/services/integrations/wix/seo.py
new file mode 100644
index 00000000..febf48c4
--- /dev/null
+++ b/backend/services/integrations/wix/seo.py
@@ -0,0 +1,300 @@
+"""
+SEO Data Builder for Wix Blog Posts
+
+Builds Wix-compatible seoData objects from ALwrity SEO metadata.
+"""
+
+from typing import Dict, Any, Optional
+from loguru import logger
+
+
+def build_seo_data(seo_metadata: Dict[str, Any], default_title: str = None) -> Optional[Dict[str, Any]]:
+    """
+    Build Wix seoData object from our SEO metadata format.
+    
+    Args:
+        seo_metadata: SEO metadata dict with fields like:
+            - seo_title: SEO optimized title
+            - meta_description: Meta description
+            - focus_keyword: Main keyword
+            - blog_tags: List of tag strings (for keywords)
+            - open_graph: Open Graph data dict
+            - canonical_url: Canonical URL
+        default_title: Fallback title if seo_title not provided
+        
+    Returns:
+        Wix seoData object with settings.keywords and tags array, or None if empty
+    """
+    seo_data = {
+        'settings': {
+            'keywords': []
+        },
+        'tags': []
+    }
+    
+    # Build keywords array
+    keywords_list = []
+    
+    # Add main keyword (focus_keyword) if provided
+    focus_keyword = seo_metadata.get('focus_keyword')
+    if focus_keyword:
+        keywords_list.append({
+            'term': str(focus_keyword),
+            'isMain': True
+        })
+    
+    # Add additional keywords from blog_tags or other sources
+    blog_tags = seo_metadata.get('blog_tags', [])
+    if isinstance(blog_tags, list):
+        for tag in blog_tags:
+            tag_str = str(tag).strip()
+            if tag_str and tag_str != focus_keyword:  # Don't duplicate main keyword
+                keywords_list.append({
+                    'term': tag_str,
+                    'isMain': False
+                })
+    
+    # Add social hashtags as keywords if available
+    social_hashtags = seo_metadata.get('social_hashtags', [])
+    if isinstance(social_hashtags, list):
+        for hashtag in social_hashtags:
+            # Remove # if present
+            hashtag_str = str(hashtag).strip().lstrip('#')
+            if hashtag_str and hashtag_str != focus_keyword:
+                keywords_list.append({
+                    'term': hashtag_str,
+                    'isMain': False
+                })
+    
+    seo_data['settings']['keywords'] = keywords_list
+    
+    # Validate keywords list is not empty (or ensure at least one keyword exists)
+    if not seo_data['settings']['keywords']:
+        logger.warning("No keywords found in SEO metadata, adding empty keywords array")
+    
+    # Build tags array (meta tags, Open Graph, etc.)
+    tags_list = []
+    
+    # Meta description
+    meta_description = seo_metadata.get('meta_description')
+    if meta_description:
+        tags_list.append({
+            'type': 'meta',
+            'props': {
+                'name': 'description',
+                'content': str(meta_description)
+            },
+            'custom': True,
+            'disabled': False
+        })
+    
+    # SEO title - 'title' type uses 'children' field, not 'props.content'
+    seo_title = seo_metadata.get('seo_title') or default_title
+    if seo_title:
+        tags_list.append({
+            'type': 'title',
+            'children': str(seo_title),  # Title tags use 'children', not 'props.content'
+            'custom': True,
+            'disabled': False
+        })
+    
+    # Open Graph tags
+    open_graph = seo_metadata.get('open_graph', {})
+    if isinstance(open_graph, dict):
+        # OG Title
+        og_title = open_graph.get('title') or seo_title
+        if og_title:
+            tags_list.append({
+                'type': 'meta',
+                'props': {
+                    'property': 'og:title',
+                    'content': str(og_title)
+                },
+                'custom': True,
+                'disabled': False
+            })
+        
+        # OG Description
+        og_description = open_graph.get('description') or meta_description
+        if og_description:
+            tags_list.append({
+                'type': 'meta',
+                'props': {
+                    'property': 'og:description',
+                    'content': str(og_description)
+                },
+                'custom': True,
+                'disabled': False
+            })
+        
+        # OG Image
+        og_image = open_graph.get('image')
+        if og_image:
+            # Skip base64 images for OG tags (Wix needs URLs)
+            if isinstance(og_image, str) and (og_image.startswith('http://') or og_image.startswith('https://')):
+                tags_list.append({
+                    'type': 'meta',
+                    'props': {
+                        'property': 'og:image',
+                        'content': og_image
+                    },
+                    'custom': True,
+                    'disabled': False
+                })
+        
+        # OG Type
+        tags_list.append({
+            'type': 'meta',
+            'props': {
+                'property': 'og:type',
+                'content': 'article'
+            },
+            'custom': True,
+            'disabled': False
+        })
+        
+        # OG URL (canonical or provided URL)
+        og_url = open_graph.get('url') or seo_metadata.get('canonical_url')
+        if og_url:
+            tags_list.append({
+                'type': 'meta',
+                'props': {
+                    'property': 'og:url',
+                    'content': str(og_url)
+                },
+                'custom': True,
+                'disabled': False
+            })
+    
+    # Twitter Card tags
+    twitter_card = seo_metadata.get('twitter_card', {})
+    if isinstance(twitter_card, dict):
+        twitter_title = twitter_card.get('title') or seo_title
+        if twitter_title:
+            tags_list.append({
+                'type': 'meta',
+                'props': {
+                    'name': 'twitter:title',
+                    'content': str(twitter_title)
+                },
+                'custom': True,
+                'disabled': False
+            })
+        
+        twitter_description = twitter_card.get('description') or meta_description
+        if twitter_description:
+            tags_list.append({
+                'type': 'meta',
+                'props': {
+                    'name': 'twitter:description',
+                    'content': str(twitter_description)
+                },
+                'custom': True,
+                'disabled': False
+            })
+        
+        twitter_image = twitter_card.get('image')
+        if twitter_image and isinstance(twitter_image, str) and (twitter_image.startswith('http://') or twitter_image.startswith('https://')):
+            tags_list.append({
+                'type': 'meta',
+                'props': {
+                    'name': 'twitter:image',
+                    'content': twitter_image
+                },
+                'custom': True,
+                'disabled': False
+            })
+        
+        twitter_card_type = twitter_card.get('card', 'summary_large_image')
+        tags_list.append({
+            'type': 'meta',
+            'props': {
+                'name': 'twitter:card',
+                'content': str(twitter_card_type)
+            },
+            'custom': True,
+            'disabled': False
+        })
+    
+    # Canonical URL as link tag
+    canonical_url = seo_metadata.get('canonical_url')
+    if canonical_url:
+        tags_list.append({
+            'type': 'link',
+            'props': {
+                'rel': 'canonical',
+                'href': str(canonical_url)
+            },
+            'custom': True,
+            'disabled': False
+        })
+    
+    # Validate all tags have required fields before adding
+    validated_tags = []
+    for tag in tags_list:
+        if not isinstance(tag, dict):
+            logger.warning(f"Skipping invalid tag (not a dict): {type(tag)}")
+            continue
+        # Ensure required fields exist
+        if 'type' not in tag:
+            logger.warning("Skipping tag missing 'type' field")
+            continue
+        # Ensure 'custom' and 'disabled' fields exist
+        if 'custom' not in tag:
+            tag['custom'] = True
+        if 'disabled' not in tag:
+            tag['disabled'] = False
+        # Validate tag structure based on type
+        tag_type = tag.get('type')
+        if tag_type == 'title':
+            if 'children' not in tag or not tag['children']:
+                logger.warning("Skipping title tag with missing/invalid 'children' field")
+                continue
+        elif tag_type == 'meta':
+            if 'props' not in tag or not isinstance(tag['props'], dict):
+                logger.warning("Skipping meta tag with missing/invalid 'props' field")
+                continue
+            if 'name' not in tag['props'] and 'property' not in tag['props']:
+                logger.warning("Skipping meta tag with missing 'name' or 'property' in props")
+                continue
+            # Ensure 'content' exists and is not empty
+            if 'content' not in tag['props'] or not str(tag['props'].get('content', '')).strip():
+                logger.warning(f"Skipping meta tag with missing/empty 'content': {tag.get('props', {})}")
+                continue
+        elif tag_type == 'link':
+            if 'props' not in tag or not isinstance(tag['props'], dict):
+                logger.warning("Skipping link tag with missing/invalid 'props' field")
+                continue
+            # Ensure 'href' exists and is not empty for link tags
+            if 'href' not in tag['props'] or not str(tag['props'].get('href', '')).strip():
+                logger.warning(f"Skipping link tag with missing/empty 'href': {tag.get('props', {})}")
+                continue
+        validated_tags.append(tag)
+    
+    seo_data['tags'] = validated_tags
+    
+    # Final validation: ensure seoData structure is complete
+    if not isinstance(seo_data['settings'], dict):
+        logger.error("seoData.settings is not a dict, creating default")
+        seo_data['settings'] = {'keywords': []}
+    if not isinstance(seo_data['settings'].get('keywords'), list):
+        logger.error("seoData.settings.keywords is not a list, creating empty list")
+        seo_data['settings']['keywords'] = []
+    if not isinstance(seo_data['tags'], list):
+        logger.error("seoData.tags is not a list, creating empty list")
+        seo_data['tags'] = []
+    
+    # CRITICAL: Per Wix API patterns, omit empty structures instead of including them as {}
+    # If keywords is empty, omit settings entirely
+    if not seo_data['settings'].get('keywords'):
+        logger.debug("No keywords found, omitting settings from seoData")
+        seo_data.pop('settings', None)
+    
+    logger.debug(f"Built SEO data: {len(validated_tags)} tags, {len(keywords_list)} keywords")
+    
+    # Only return seoData if we have at least keywords or tags
+    if keywords_list or validated_tags:
+        return seo_data
+    
+    return None
+
diff --git a/backend/services/llm_providers/main_text_generation.py b/backend/services/llm_providers/main_text_generation.py
index e7fccae6..01333265 100644
--- a/backend/services/llm_providers/main_text_generation.py
+++ b/backend/services/llm_providers/main_text_generation.py
@@ -9,6 +9,7 @@ import json
 from typing import Optional, Dict, Any
 from datetime import datetime
 from loguru import logger
+from fastapi import HTTPException
 from ..onboarding.api_key_manager import APIKeyManager
 
 from .gemini_provider import gemini_text_response, gemini_structured_json_response
@@ -129,11 +130,16 @@ def llm_text_gen(prompt: str, system_prompt: Optional[str] = None, json_struct:
                 pricing_service = PricingService(db)
                 
                 # Estimate tokens from prompt (input tokens)
-                # Note: We estimate output tokens conservatively (assume response is similar length to prompt)
-                # This prevents underestimating total token usage
+                # CRITICAL: Use worst-case scenario (input + max_tokens) for validation to prevent abuse
+                # This ensures we block requests that would exceed limits even if response is longer than expected
                 input_tokens = int(len(prompt.split()) * 1.3)
-                # Conservative estimate: assume output tokens ≈ input tokens * 1.0 (can be up to max_tokens)
-                estimated_output_tokens = min(input_tokens, max_tokens) if max_tokens else int(input_tokens * 0.8)
+                # Worst-case estimate: assume maximum possible output tokens (max_tokens if specified)
+                # This prevents abuse where actual response tokens exceed the estimate
+                if max_tokens:
+                    estimated_output_tokens = max_tokens  # Use maximum allowed output tokens
+                else:
+                    # If max_tokens not specified, use conservative estimate (input * 1.5)
+                    estimated_output_tokens = int(input_tokens * 1.5)
                 estimated_total_tokens = input_tokens + estimated_output_tokens
                 
                 # Check limits using sync method from pricing service (strict enforcement)
@@ -146,7 +152,14 @@ def llm_text_gen(prompt: str, system_prompt: Optional[str] = None, json_struct:
                 
                 if not can_proceed:
                     logger.warning(f"[llm_text_gen] Subscription limit exceeded for user {user_id}: {message}")
-                    raise RuntimeError(f"Subscription limit exceeded: {message}")
+                    # Raise HTTPException(429) with usage info so frontend can display subscription modal
+                    error_detail = {
+                        'error': message,
+                        'message': message,
+                        'provider': actual_provider_name or provider_enum.value,
+                        'usage_info': usage_info if usage_info else {}
+                    }
+                    raise HTTPException(status_code=429, detail=error_detail)
                 
                 # Get current usage for limit checking only
                 current_period = pricing_service.get_current_billing_period(user_id) or datetime.now().strftime("%Y-%m")
@@ -159,6 +172,9 @@ def llm_text_gen(prompt: str, system_prompt: Optional[str] = None, json_struct:
                 
             finally:
                 db.close()
+        except HTTPException:
+            # Re-raise HTTPExceptions (e.g., 429 subscription limit) - preserve error details
+            raise
         except RuntimeError:
             # Re-raise subscription limit errors
             raise
@@ -244,7 +260,8 @@ def llm_text_gen(prompt: str, system_prompt: Optional[str] = None, json_struct:
                     db_track = next(get_db())
                     try:
                         # Estimate tokens from prompt and response
-                        tokens_input = estimated_tokens  # Already calculated above
+                        # Recalculate input tokens from prompt (consistent with pre-flight estimation)
+                        tokens_input = int(len(prompt.split()) * 1.3)
                         tokens_output = int(len(str(response_text).split()) * 1.3)  # Estimate output tokens
                         tokens_total = tokens_input + tokens_output
                         
@@ -259,45 +276,186 @@ def llm_text_gen(prompt: str, system_prompt: Optional[str] = None, json_struct:
                         
                         logger.debug(f"[llm_text_gen] Looking for usage summary: user_id={user_id}, period={current_period}")
                         
+                        # Get limits once for safety check (to prevent exceeding limits even if actual usage > estimate)
+                        provider_name = provider_enum.value
+                        limits = pricing.get_user_limits(user_id)
+                        token_limit = 0
+                        if limits and limits.get('limits'):
+                            token_limit = limits['limits'].get(f"{provider_name}_tokens", 0) or 0
+                        
+                        # CRITICAL: Use raw SQL to read current values directly from DB, bypassing SQLAlchemy cache
+                        # This ensures we always get the absolute latest committed values, even across different sessions
+                        from sqlalchemy import text
+                        current_calls_before = 0
+                        current_tokens_before = 0
+                        record_count = 0  # Initialize to ensure it's always defined
+                        
+                        # CRITICAL: First check if record exists using COUNT query
+                        try:
+                            check_query = text("SELECT COUNT(*) FROM usage_summaries WHERE user_id = :user_id AND billing_period = :period")
+                            record_count = db_track.execute(check_query, {'user_id': user_id, 'period': current_period}).scalar()
+                            logger.debug(f"[llm_text_gen] 🔍 DEBUG: Record count check - found {record_count} record(s) for user={user_id}, period={current_period}")
+                        except Exception as count_error:
+                            logger.error(f"[llm_text_gen] ❌ COUNT query failed: {count_error}", exc_info=True)
+                            record_count = 0
+                        
+                        if record_count and record_count > 0:
+                            # Record exists - read current values with raw SQL
+                            try:
+                                # Validate provider_name to prevent SQL injection (whitelist approach)
+                                valid_providers = ['gemini', 'openai', 'anthropic', 'mistral']
+                                if provider_name not in valid_providers:
+                                    raise ValueError(f"Invalid provider_name for SQL query: {provider_name}")
+                                
+                                # Read current values directly from database using raw SQL
+                                # CRITICAL: This bypasses SQLAlchemy's session cache and gets absolute latest values
+                                sql_query = text(f"""
+                                    SELECT {provider_name}_calls, {provider_name}_tokens 
+                                    FROM usage_summaries 
+                                    WHERE user_id = :user_id AND billing_period = :period 
+                                    LIMIT 1
+                                """)
+                                logger.debug(f"[llm_text_gen] 🔍 Executing raw SQL for EXISTING record: SELECT {provider_name}_calls, {provider_name}_tokens WHERE user_id={user_id}, period={current_period}")
+                                result = db_track.execute(sql_query, {'user_id': user_id, 'period': current_period}).first()
+                                if result:
+                                    raw_calls = result[0] if result[0] is not None else 0
+                                    raw_tokens = result[1] if result[1] is not None else 0
+                                    current_calls_before = raw_calls
+                                    current_tokens_before = raw_tokens
+                                    logger.debug(f"[llm_text_gen] ✅ Raw SQL SUCCESS: Found EXISTING record - calls={current_calls_before}, tokens={current_tokens_before} (provider={provider_name}, column={provider_name}_calls/{provider_name}_tokens)")
+                                    logger.debug(f"[llm_text_gen] 🔍 Raw SQL returned row: {result}, extracted calls={raw_calls}, tokens={raw_tokens}")
+                                else:
+                                    logger.error(f"[llm_text_gen] ❌ CRITICAL BUG: Record EXISTS (count={record_count}) but SELECT query returned None! Query: {sql_query}")
+                                    # Fallback: Use ORM to get values
+                                    summary_fallback = db_track.query(UsageSummary).filter(
+                                        UsageSummary.user_id == user_id,
+                                        UsageSummary.billing_period == current_period
+                                    ).first()
+                                    if summary_fallback:
+                                        db_track.refresh(summary_fallback)
+                                        current_calls_before = getattr(summary_fallback, f"{provider_name}_calls", 0) or 0
+                                        current_tokens_before = getattr(summary_fallback, f"{provider_name}_tokens", 0) or 0
+                                        logger.warning(f"[llm_text_gen] ⚠️ Using ORM fallback: calls={current_calls_before}, tokens={current_tokens_before}")
+                            except Exception as sql_error:
+                                logger.error(f"[llm_text_gen] ❌ Raw SQL query failed: {sql_error}", exc_info=True)
+                                # Fallback: Use ORM to get values
+                                summary_fallback = db_track.query(UsageSummary).filter(
+                                    UsageSummary.user_id == user_id,
+                                    UsageSummary.billing_period == current_period
+                                ).first()
+                                if summary_fallback:
+                                    db_track.refresh(summary_fallback)
+                                    current_calls_before = getattr(summary_fallback, f"{provider_name}_calls", 0) or 0
+                                    current_tokens_before = getattr(summary_fallback, f"{provider_name}_tokens", 0) or 0
+                        else:
+                            logger.debug(f"[llm_text_gen] ℹ️ No record exists yet (will create new) - user={user_id}, period={current_period}")
+                        
+                        # Get or create usage summary object (needed for ORM update)
                         summary = db_track.query(UsageSummary).filter(
                             UsageSummary.user_id == user_id,
                             UsageSummary.billing_period == current_period
                         ).first()
                         
                         if not summary:
-                            logger.info(f"[llm_text_gen] Creating new usage summary for user {user_id}, period {current_period}")
+                            logger.debug(f"[llm_text_gen] Creating NEW usage summary for user {user_id}, period {current_period}")
                             summary = UsageSummary(
                                 user_id=user_id,
                                 billing_period=current_period
                             )
                             db_track.add(summary)
                             db_track.flush()  # Ensure summary is persisted before updating
+                            # New record - values are already 0, no need to set
+                            logger.debug(f"[llm_text_gen] ✅ New summary created - starting from 0")
+                        else:
+                            # CRITICAL: Update the ORM object with values from raw SQL query
+                            # This ensures the ORM object reflects the actual database state before we increment
+                            logger.debug(f"[llm_text_gen] 🔄 Existing summary found - syncing with raw SQL values: calls={current_calls_before}, tokens={current_tokens_before}")
+                            setattr(summary, f"{provider_name}_calls", current_calls_before)
+                            if provider_enum in [APIProvider.GEMINI, APIProvider.OPENAI, APIProvider.ANTHROPIC, APIProvider.MISTRAL]:
+                                setattr(summary, f"{provider_name}_tokens", current_tokens_before)
+                            logger.debug(f"[llm_text_gen] ✅ Synchronized ORM object: {provider_name}_calls={current_calls_before}, {provider_name}_tokens={current_tokens_before}")
                         
-                        # Get "before" state for unified log
-                        provider_name = provider_enum.value
-                        current_calls_before = getattr(summary, f"{provider_name}_calls", 0) or 0
+                        logger.debug(f"[llm_text_gen] Current {provider_name}_calls from DB (raw SQL): {current_calls_before}")
                         
                         # Update provider-specific counters (sync operation)
                         new_calls = current_calls_before + 1
-                        setattr(summary, f"{provider_name}_calls", new_calls)
-                        logger.debug(f"[llm_text_gen] Updated {provider_name}_calls: {current_calls_before} -> {new_calls}")
                         
-                        # Update token usage for LLM providers
+                        # CRITICAL: Use direct SQL UPDATE instead of ORM setattr for dynamic attributes
+                        # SQLAlchemy doesn't detect changes when using setattr() on dynamic attributes
+                        # Using raw SQL UPDATE ensures the change is persisted
+                        from sqlalchemy import text
+                        update_calls_query = text(f"""
+                            UPDATE usage_summaries 
+                            SET {provider_name}_calls = :new_calls 
+                            WHERE user_id = :user_id AND billing_period = :period
+                        """)
+                        db_track.execute(update_calls_query, {
+                            'new_calls': new_calls,
+                            'user_id': user_id,
+                            'period': current_period
+                        })
+                        logger.debug(f"[llm_text_gen] Updated {provider_name}_calls via SQL: {current_calls_before} -> {new_calls}")
+                        
+                        # Update token usage for LLM providers with safety check
+                        # CRITICAL: Use current_tokens_before from raw SQL query (NOT from ORM object)
+                        # The ORM object may have stale values, but raw SQL always has the latest committed values
                         if provider_enum in [APIProvider.GEMINI, APIProvider.OPENAI, APIProvider.ANTHROPIC, APIProvider.MISTRAL]:
-                            current_tokens_before = getattr(summary, f"{provider_name}_tokens", 0) or 0
-                            new_tokens = current_tokens_before + tokens_total
-                            setattr(summary, f"{provider_name}_tokens", new_tokens)
-                            logger.debug(f"[llm_text_gen] Updated {provider_name}_tokens: {current_tokens_before} -> {new_tokens}")
+                            logger.debug(f"[llm_text_gen] Current {provider_name}_tokens from DB (raw SQL): {current_tokens_before}")
+                            
+                            # SAFETY CHECK: Prevent exceeding token limit even if actual usage exceeds estimate
+                            # This prevents abuse where actual response tokens exceed pre-flight validation estimate
+                            projected_new_tokens = current_tokens_before + tokens_total
+                            
+                            # If limit is set (> 0) and would be exceeded, cap at limit
+                            if token_limit > 0 and projected_new_tokens > token_limit:
+                                logger.warning(
+                                    f"[llm_text_gen] ⚠️ ACTUAL token usage ({tokens_total}) exceeded estimate. "
+                                    f"Would exceed limit: {projected_new_tokens} > {token_limit}. "
+                                    f"Capping tracked tokens at limit to prevent abuse."
+                                )
+                                # Cap at limit to prevent abuse
+                                new_tokens = token_limit
+                                # Adjust tokens_total for accurate total tracking
+                                tokens_total = token_limit - current_tokens_before
+                                if tokens_total < 0:
+                                    tokens_total = 0
+                            else:
+                                new_tokens = projected_new_tokens
+                            
+                            # CRITICAL: Use direct SQL UPDATE instead of ORM setattr for dynamic attributes
+                            update_tokens_query = text(f"""
+                                UPDATE usage_summaries 
+                                SET {provider_name}_tokens = :new_tokens 
+                                WHERE user_id = :user_id AND billing_period = :period
+                            """)
+                            db_track.execute(update_tokens_query, {
+                                'new_tokens': new_tokens,
+                                'user_id': user_id,
+                                'period': current_period
+                            })
+                            logger.debug(f"[llm_text_gen] Updated {provider_name}_tokens via SQL: {current_tokens_before} -> {new_tokens}")
                         else:
                             current_tokens_before = 0
                             new_tokens = 0
                         
-                        # Update totals
+                        # Update totals using SQL UPDATE
                         old_total_calls = summary.total_calls or 0
                         old_total_tokens = summary.total_tokens or 0
-                        summary.total_calls = old_total_calls + 1
-                        summary.total_tokens = old_total_tokens + tokens_total
-                        logger.debug(f"[llm_text_gen] Updated totals: calls {old_total_calls} -> {summary.total_calls}, tokens {old_total_tokens} -> {summary.total_tokens}")
+                        new_total_calls = old_total_calls + 1
+                        new_total_tokens = old_total_tokens + tokens_total
+                        
+                        update_totals_query = text("""
+                            UPDATE usage_summaries 
+                            SET total_calls = :total_calls, total_tokens = :total_tokens 
+                            WHERE user_id = :user_id AND billing_period = :period
+                        """)
+                        db_track.execute(update_totals_query, {
+                            'total_calls': new_total_calls,
+                            'total_tokens': new_total_tokens,
+                            'user_id': user_id,
+                            'period': current_period
+                        })
+                        logger.debug(f"[llm_text_gen] Updated totals via SQL: calls {old_total_calls} -> {new_total_calls}, tokens {old_total_tokens} -> {new_total_tokens}")
                         
                         # Get plan details for unified log
                         limits = pricing.get_user_limits(user_id)
@@ -310,12 +468,52 @@ def llm_text_gen(prompt: str, system_prompt: Optional[str] = None, json_struct:
                         current_images_before = getattr(summary, "stability_calls", 0) or 0
                         image_limit = limits['limits'].get("stability_calls", 0) if limits else 0
                         
-                        db_track.commit()
-                        logger.info(f"[llm_text_gen] ✅ Successfully tracked usage: user {user_id} -> provider {provider_name} -> {new_calls} calls, {new_tokens} tokens")
+                        # CRITICAL DEBUG: Print diagnostic info BEFORE commit (always visible, flushed immediately)
+                        import sys
+                        debug_msg = f"[DEBUG] BEFORE COMMIT - Record count: {record_count}, Raw SQL values: calls={current_calls_before}, tokens={current_tokens_before}, Provider: {provider_name}, Period: {current_period}, New calls will be: {new_calls}, New tokens will be: {new_tokens}"
+                        print(debug_msg, flush=True)
+                        sys.stdout.flush()
+                        logger.debug(f"[llm_text_gen] {debug_msg}")
+                        
+                        # CRITICAL: Flush before commit to ensure changes are immediately visible to other sessions
+                        db_track.flush()  # Flush to ensure changes are in DB (not just in transaction)
+                        db_track.commit()  # Commit transaction to make changes visible to other sessions
+                        logger.debug(f"[llm_text_gen] ✅ Successfully tracked usage: user {user_id} -> provider {provider_name} -> {new_calls} calls, {new_tokens} tokens (COMMITTED to DB)")
+                        logger.debug(f"[llm_text_gen] Database state after commit: {provider_name}_calls={new_calls}, {provider_name}_tokens={new_tokens} (should be visible to next session)")
+                        
+                        # CRITICAL: Verify commit worked by reading back from DB immediately after commit
+                        try:
+                            verify_query = text(f"SELECT {provider_name}_calls, {provider_name}_tokens FROM usage_summaries WHERE user_id = :user_id AND billing_period = :period LIMIT 1")
+                            verify_result = db_track.execute(verify_query, {'user_id': user_id, 'period': current_period}).first()
+                            if verify_result:
+                                verified_calls = verify_result[0] if verify_result[0] is not None else 0
+                                verified_tokens = verify_result[1] if verify_result[1] is not None else 0
+                                logger.debug(f"[llm_text_gen] ✅ VERIFICATION AFTER COMMIT: Read back calls={verified_calls}, tokens={verified_tokens} (expected: calls={new_calls}, tokens={new_tokens})")
+                                if verified_calls != new_calls or verified_tokens != new_tokens:
+                                    logger.error(f"[llm_text_gen] ❌ CRITICAL: COMMIT VERIFICATION FAILED! Expected calls={new_calls}, tokens={new_tokens}, but DB has calls={verified_calls}, tokens={verified_tokens}")
+                                    # Force another commit attempt
+                                    db_track.commit()
+                                    verify_result2 = db_track.execute(verify_query, {'user_id': user_id, 'period': current_period}).first()
+                                    if verify_result2:
+                                        verified_calls2 = verify_result2[0] if verify_result2[0] is not None else 0
+                                        verified_tokens2 = verify_result2[1] if verify_result2[1] is not None else 0
+                                        logger.debug(f"[llm_text_gen] 🔄 After second commit attempt: calls={verified_calls2}, tokens={verified_tokens2}")
+                                else:
+                                    logger.debug(f"[llm_text_gen] ✅ COMMIT VERIFICATION PASSED: Values match expected values")
+                            else:
+                                logger.error(f"[llm_text_gen] ❌ CRITICAL: COMMIT VERIFICATION FAILED! Record not found after commit!")
+                        except Exception as verify_error:
+                            logger.error(f"[llm_text_gen] ❌ Error verifying commit: {verify_error}", exc_info=True)
                         
                         # UNIFIED SUBSCRIPTION LOG - Shows before/after state in one message
                         # Use actual_provider_name (e.g., "huggingface") instead of enum value (e.g., "mistral")
                         # Include image stats in the log
+                        # DEBUG: Log the actual values being used
+                        logger.debug(f"[llm_text_gen] 📊 FINAL VALUES FOR LOG: calls_before={current_calls_before}, calls_after={new_calls}, tokens_before={current_tokens_before}, tokens_after={new_tokens}, provider={provider_name}, enum={provider_enum}")
+                        
+                        # CRITICAL DEBUG: Print diagnostic info to stdout (always visible)
+                        print(f"[DEBUG] Record count: {record_count}, Raw SQL values: calls={current_calls_before}, tokens={current_tokens_before}, Provider: {provider_name}")
+                        
                         print(f"""
 [SUBSCRIPTION] LLM Text Generation
 ├─ User: {user_id}
@@ -407,7 +605,8 @@ def llm_text_gen(prompt: str, system_prompt: Optional[str] = None, json_struct:
                             db_track = next(get_db())
                             try:
                                 # Estimate tokens from prompt and response
-                                tokens_input = estimated_tokens
+                                # Recalculate input tokens from prompt (consistent with pre-flight estimation)
+                                tokens_input = int(len(prompt.split()) * 1.3)
                                 tokens_output = int(len(str(response_text).split()) * 1.3)
                                 tokens_total = tokens_input + tokens_output
                                 
@@ -418,6 +617,49 @@ def llm_text_gen(prompt: str, system_prompt: Optional[str] = None, json_struct:
                                 pricing = PricingService(db_track)
                                 current_period = pricing.get_current_billing_period(user_id) or datetime.now().strftime("%Y-%m")
                                 
+                                # Get limits once for safety check (to prevent exceeding limits even if actual usage > estimate)
+                                provider_name = provider_enum.value
+                                limits = pricing.get_user_limits(user_id)
+                                token_limit = 0
+                                if limits and limits.get('limits'):
+                                    token_limit = limits['limits'].get(f"{provider_name}_tokens", 0) or 0
+                                
+                                # CRITICAL: Use raw SQL to read current values directly from DB, bypassing SQLAlchemy cache
+                                from sqlalchemy import text
+                                current_calls_before = 0
+                                current_tokens_before = 0
+                                
+                                try:
+                                    # Validate provider_name to prevent SQL injection
+                                    valid_providers = ['gemini', 'openai', 'anthropic', 'mistral']
+                                    if provider_name not in valid_providers:
+                                        raise ValueError(f"Invalid provider_name for SQL query: {provider_name}")
+                                    
+                                    # Read current values directly from database using raw SQL
+                                    sql_query = text(f"""
+                                        SELECT {provider_name}_calls, {provider_name}_tokens 
+                                        FROM usage_summaries 
+                                        WHERE user_id = :user_id AND billing_period = :period 
+                                        LIMIT 1
+                                    """)
+                                    result = db_track.execute(sql_query, {'user_id': user_id, 'period': current_period}).first()
+                                    if result:
+                                        current_calls_before = result[0] if result[0] is not None else 0
+                                        current_tokens_before = result[1] if result[1] is not None else 0
+                                        logger.debug(f"[llm_text_gen] Raw SQL read current values (fallback): calls={current_calls_before}, tokens={current_tokens_before}")
+                                except Exception as sql_error:
+                                    logger.warning(f"[llm_text_gen] Raw SQL query failed (fallback), falling back to ORM: {sql_error}")
+                                    # Fallback to ORM query if raw SQL fails
+                                    summary = db_track.query(UsageSummary).filter(
+                                        UsageSummary.user_id == user_id,
+                                        UsageSummary.billing_period == current_period
+                                    ).first()
+                                    if summary:
+                                        db_track.refresh(summary)
+                                        current_calls_before = getattr(summary, f"{provider_name}_calls", 0) or 0
+                                        current_tokens_before = getattr(summary, f"{provider_name}_tokens", 0) or 0
+                                
+                                # Get or create usage summary object (needed for ORM update)
                                 summary = db_track.query(UsageSummary).filter(
                                     UsageSummary.user_id == user_id,
                                     UsageSummary.billing_period == current_period
@@ -430,41 +672,68 @@ def llm_text_gen(prompt: str, system_prompt: Optional[str] = None, json_struct:
                                     )
                                     db_track.add(summary)
                                     db_track.flush()  # Ensure summary is persisted before updating
+                                else:
+                                    # CRITICAL: Update the ORM object with values from raw SQL query
+                                    # This ensures the ORM object reflects the actual database state before we increment
+                                    setattr(summary, f"{provider_name}_calls", current_calls_before)
+                                    if provider_enum in [APIProvider.GEMINI, APIProvider.OPENAI, APIProvider.ANTHROPIC, APIProvider.MISTRAL]:
+                                        setattr(summary, f"{provider_name}_tokens", current_tokens_before)
+                                    logger.debug(f"[llm_text_gen] Synchronized summary object with raw SQL values (fallback): calls={current_calls_before}, tokens={current_tokens_before}")
                                 
-                                # Get "before" state for unified log
-                                provider_name = provider_enum.value
-                                current_calls_before = getattr(summary, f"{provider_name}_calls", 0) or 0
+                                # Get "before" state for unified log (from raw SQL query)
+                                logger.debug(f"[llm_text_gen] Current {provider_name}_calls from DB (fallback, raw SQL): {current_calls_before}")
                                 
                                 # Update provider-specific counters (sync operation)
                                 new_calls = current_calls_before + 1
                                 setattr(summary, f"{provider_name}_calls", new_calls)
                                 
-                                # Update token usage for LLM providers
+                                # Update token usage for LLM providers with safety check
+                                # Use current_tokens_before from raw SQL query (most reliable)
                                 if provider_enum in [APIProvider.GEMINI, APIProvider.OPENAI, APIProvider.ANTHROPIC, APIProvider.MISTRAL]:
-                                    current_tokens_before = getattr(summary, f"{provider_name}_tokens", 0) or 0
-                                    new_tokens = current_tokens_before + tokens_total
+                                    logger.debug(f"[llm_text_gen] Current {provider_name}_tokens from DB (fallback, raw SQL): {current_tokens_before}")
+                                    
+                                    # SAFETY CHECK: Prevent exceeding token limit even if actual usage exceeds estimate
+                                    # This prevents abuse where actual response tokens exceed pre-flight validation estimate
+                                    projected_new_tokens = current_tokens_before + tokens_total
+                                    
+                                    # If limit is set (> 0) and would be exceeded, cap at limit
+                                    if token_limit > 0 and projected_new_tokens > token_limit:
+                                        logger.warning(
+                                            f"[llm_text_gen] ⚠️ ACTUAL token usage ({tokens_total}) exceeded estimate in fallback provider. "
+                                            f"Would exceed limit: {projected_new_tokens} > {token_limit}. "
+                                            f"Capping tracked tokens at limit to prevent abuse."
+                                        )
+                                        # Cap at limit to prevent abuse
+                                        new_tokens = token_limit
+                                        # Adjust tokens_total for accurate total tracking
+                                        tokens_total = token_limit - current_tokens_before
+                                        if tokens_total < 0:
+                                            tokens_total = 0
+                                    else:
+                                        new_tokens = projected_new_tokens
+                                    
                                     setattr(summary, f"{provider_name}_tokens", new_tokens)
                                 else:
                                     current_tokens_before = 0
                                     new_tokens = 0
                                 
-                                # Update totals
+                                # Update totals (using potentially capped tokens_total from safety check)
                                 summary.total_calls = (summary.total_calls or 0) + 1
                                 summary.total_tokens = (summary.total_tokens or 0) + tokens_total
                                 
-                                # Get plan details for unified log
-                                limits = pricing.get_user_limits(user_id)
+                                # Get plan details for unified log (limits already retrieved above)
                                 plan_name = limits.get('plan_name', 'unknown') if limits else 'unknown'
                                 tier = limits.get('tier', 'unknown') if limits else 'unknown'
                                 call_limit = limits['limits'].get(f"{provider_name}_calls", 0) if limits else 0
-                                token_limit = limits['limits'].get(f"{provider_name}_tokens", 0) if limits else 0
                                 
                                 # Get image stats for unified log
                                 current_images_before = getattr(summary, "stability_calls", 0) or 0
                                 image_limit = limits['limits'].get("stability_calls", 0) if limits else 0
                                 
-                                db_track.commit()
-                                logger.info(f"[llm_text_gen] ✅ Successfully tracked fallback usage: user {user_id} -> provider {provider_name} -> {new_calls} calls, {new_tokens} tokens")
+                                # CRITICAL: Flush before commit to ensure changes are immediately visible to other sessions
+                                db_track.flush()  # Flush to ensure changes are in DB (not just in transaction)
+                                db_track.commit()  # Commit transaction to make changes visible to other sessions
+                                logger.info(f"[llm_text_gen] ✅ Successfully tracked fallback usage: user {user_id} -> provider {provider_name} -> {new_calls} calls, {new_tokens} tokens (committed)")
                                 
                                 # UNIFIED SUBSCRIPTION LOG for fallback
                                 # Use actual_provider_name (e.g., "huggingface") instead of enum value (e.g., "mistral")
diff --git a/backend/services/subscription/limit_validation.py b/backend/services/subscription/limit_validation.py
new file mode 100644
index 00000000..da4c678d
--- /dev/null
+++ b/backend/services/subscription/limit_validation.py
@@ -0,0 +1,725 @@
+"""
+Limit Validation Module
+Handles subscription limit checking and validation logic.
+Extracted from pricing_service.py for better modularity.
+"""
+
+from typing import Dict, Any, Optional, List, Tuple, TYPE_CHECKING
+from datetime import datetime, timedelta
+from sqlalchemy import text
+from loguru import logger
+
+from models.subscription_models import (
+    UserSubscription, UsageSummary, SubscriptionPlan, 
+    APIProvider, SubscriptionTier
+)
+
+if TYPE_CHECKING:
+    from .pricing_service import PricingService
+
+
+class LimitValidator:
+    """Validates subscription limits for API usage."""
+    
+    def __init__(self, pricing_service: 'PricingService'):
+        """
+        Initialize limit validator with reference to PricingService.
+        
+        Args:
+            pricing_service: Instance of PricingService to access helper methods and cache
+        """
+        self.pricing_service = pricing_service
+        self.db = pricing_service.db
+    
+    def check_usage_limits(self, user_id: str, provider: APIProvider, 
+                          tokens_requested: int = 0, actual_provider_name: Optional[str] = None) -> Tuple[bool, str, Dict[str, Any]]:
+        """Check if user can make an API call within their limits.
+        
+        Args:
+            user_id: User ID
+            provider: APIProvider enum (may be MISTRAL for HuggingFace)
+            tokens_requested: Estimated tokens for the request
+            actual_provider_name: Optional actual provider name (e.g., "huggingface" when provider is MISTRAL)
+        
+        Returns:
+            (can_proceed, error_message, usage_info)
+        """
+        try:
+            # Use actual_provider_name if provided, otherwise use enum value
+            # This fixes cases where HuggingFace maps to MISTRAL enum but should show as "huggingface" in errors
+            display_provider_name = actual_provider_name or provider.value
+            
+            logger.debug(f"[Subscription Check] Starting limit check for user {user_id}, provider {display_provider_name}, tokens {tokens_requested}")
+            
+            # Short TTL cache to reduce DB reads under sustained traffic
+            cache_key = f"{user_id}:{provider.value}"
+            now = datetime.utcnow()
+            cached = self.pricing_service._limits_cache.get(cache_key)
+            if cached and cached.get('expires_at') and cached['expires_at'] > now:
+                logger.debug(f"[Subscription Check] Using cached result for {user_id}:{provider.value}")
+                return tuple(cached['result'])  # type: ignore
+
+            # Get user subscription first to check expiration
+            subscription = self.db.query(UserSubscription).filter(
+                UserSubscription.user_id == user_id,
+                UserSubscription.is_active == True
+            ).first()
+            
+            if subscription:
+                logger.debug(f"[Subscription Check] Found subscription for user {user_id}: plan_id={subscription.plan_id}, period_end={subscription.current_period_end}")
+            else:
+                logger.debug(f"[Subscription Check] No active subscription found for user {user_id}")
+            
+            # Check subscription expiration (STRICT: deny if expired)
+            if subscription:
+                if subscription.current_period_end < now:
+                    logger.warning(f"[Subscription Check] Subscription expired for user {user_id}: period_end={subscription.current_period_end}, now={now}")
+                    # Subscription expired - check if auto_renew is enabled
+                    if not getattr(subscription, 'auto_renew', False):
+                        # Expired and no auto-renew - deny access
+                        logger.warning(f"[Subscription Check] Subscription expired for user {user_id}, auto_renew=False, denying access")
+                        result = (False, "Subscription expired. Please renew your subscription to continue using the service.", {
+                            'expired': True,
+                            'period_end': subscription.current_period_end.isoformat()
+                        })
+                        self.pricing_service._limits_cache[cache_key] = {
+                            'result': result,
+                            'expires_at': now + timedelta(seconds=30)
+                        }
+                        return result
+                    else:
+                        # Try to auto-renew
+                        if not self.pricing_service._ensure_subscription_current(subscription):
+                            # Auto-renew failed - deny access
+                            result = (False, "Subscription expired and auto-renewal failed. Please renew manually.", {
+                                'expired': True,
+                                'auto_renew_failed': True
+                            })
+                            self.pricing_service._limits_cache[cache_key] = {
+                                'result': result,
+                                'expires_at': now + timedelta(seconds=30)
+                            }
+                            return result
+
+            # Get user limits with error handling (STRICT: fail on errors)
+            # CRITICAL: Expire SQLAlchemy objects to ensure we get fresh plan data after renewal
+            try:
+                # Force expire subscription and plan objects to avoid stale cache
+                if subscription and subscription.plan_id:
+                    plan_obj = self.db.query(SubscriptionPlan).filter(SubscriptionPlan.id == subscription.plan_id).first()
+                    if plan_obj:
+                        self.db.expire(plan_obj)
+                        logger.debug(f"[Subscription Check] Expired plan object to ensure fresh limits after renewal")
+                
+                limits = self.pricing_service.get_user_limits(user_id)
+                if limits:
+                    logger.debug(f"[Subscription Check] Retrieved limits for user {user_id}: plan={limits.get('plan_name')}, tier={limits.get('tier')}")
+                    # Log token limits for debugging
+                    token_limits = limits.get('limits', {})
+                    logger.debug(f"[Subscription Check] Token limits: gemini={token_limits.get('gemini_tokens')}, mistral={token_limits.get('mistral_tokens')}, openai={token_limits.get('openai_tokens')}, anthropic={token_limits.get('anthropic_tokens')}")
+                else:
+                    logger.debug(f"[Subscription Check] No limits found for user {user_id}, checking free tier")
+            except Exception as e:
+                logger.error(f"[Subscription Check] Error getting user limits for {user_id}: {e}", exc_info=True)
+                # STRICT: Fail closed - deny request if we can't check limits
+                return False, f"Failed to retrieve subscription limits: {str(e)}", {}
+            
+            if not limits:
+                # No subscription found - check for free tier
+                free_plan = self.db.query(SubscriptionPlan).filter(
+                    SubscriptionPlan.tier == SubscriptionTier.FREE,
+                    SubscriptionPlan.is_active == True
+                ).first()
+                if free_plan:
+                    logger.info(f"[Subscription Check] Assigning free tier to user {user_id}")
+                    limits = self.pricing_service._plan_to_limits_dict(free_plan)
+                else:
+                    # No subscription and no free tier - deny access
+                    logger.warning(f"[Subscription Check] No subscription or free tier found for user {user_id}, denying access")
+                    return False, "No subscription plan found. Please subscribe to a plan.", {}
+            
+            # Get current usage for this billing period with error handling
+            # CRITICAL: Use fresh queries to avoid SQLAlchemy cache after renewal
+            try:
+                current_period = self.pricing_service.get_current_billing_period(user_id) or datetime.now().strftime("%Y-%m")
+                
+                # Expire all objects to force fresh read from DB (critical after renewal)
+                self.db.expire_all()
+                
+                # Use raw SQL query first to bypass ORM cache, fallback to ORM if SQL fails
+                usage = None
+                try:
+                    from sqlalchemy import text
+                    sql_query = text("SELECT * FROM usage_summaries WHERE user_id = :user_id AND billing_period = :period LIMIT 1")
+                    result = self.db.execute(sql_query, {'user_id': user_id, 'period': current_period}).first()
+                    if result:
+                        # Map result to UsageSummary object
+                        usage = self.db.query(UsageSummary).filter(
+                            UsageSummary.user_id == user_id,
+                            UsageSummary.billing_period == current_period
+                        ).first()
+                        if usage:
+                            self.db.refresh(usage)  # Ensure fresh data
+                except Exception as sql_error:
+                    logger.debug(f"[Subscription Check] Raw SQL query failed, using ORM: {sql_error}")
+                    # Fallback to ORM query
+                    usage = self.db.query(UsageSummary).filter(
+                        UsageSummary.user_id == user_id,
+                        UsageSummary.billing_period == current_period
+                    ).first()
+                    if usage:
+                        self.db.refresh(usage)  # Ensure fresh data
+                
+                if not usage:
+                    # First usage this period, create summary
+                    try:
+                        usage = UsageSummary(
+                            user_id=user_id,
+                            billing_period=current_period
+                        )
+                        self.db.add(usage)
+                        self.db.commit()
+                    except Exception as create_error:
+                        logger.error(f"Error creating usage summary: {create_error}")
+                        self.db.rollback()
+                        # STRICT: Fail closed on DB error
+                        return False, f"Failed to create usage summary: {str(create_error)}", {}
+            except Exception as e:
+                logger.error(f"Error getting usage summary for {user_id}: {e}")
+                self.db.rollback()
+                # STRICT: Fail closed on DB error
+                return False, f"Failed to retrieve usage summary: {str(e)}", {}
+            
+            # Check call limits with error handling
+            # NOTE: call_limit = 0 means UNLIMITED (Enterprise plans)
+            try:
+                # Use display_provider_name for error messages, but provider.value for DB queries
+                provider_name = provider.value  # For DB field names (e.g., "mistral_calls", "mistral_tokens")
+                
+                # For LLM text generation providers, check against unified total_calls limit
+                llm_providers = ['gemini', 'openai', 'anthropic', 'mistral']
+                is_llm_provider = provider_name in llm_providers
+                
+                if is_llm_provider:
+                    # Use unified AI text generation limit (total_calls across all LLM providers)
+                    ai_text_gen_limit = limits['limits'].get('ai_text_generation_calls', 0) or 0
+                    
+                    # If unified limit not set, fall back to provider-specific limit for backwards compatibility
+                    if ai_text_gen_limit == 0:
+                        ai_text_gen_limit = limits['limits'].get(f"{provider_name}_calls", 0) or 0
+                    
+                    # Calculate total LLM provider calls (sum of gemini + openai + anthropic + mistral)
+                    current_total_llm_calls = (
+                        (usage.gemini_calls or 0) +
+                        (usage.openai_calls or 0) +
+                        (usage.anthropic_calls or 0) +
+                        (usage.mistral_calls or 0)
+                    )
+                    
+                    # Only enforce limit if limit > 0 (0 means unlimited for Enterprise)
+                    if ai_text_gen_limit > 0 and current_total_llm_calls >= ai_text_gen_limit:
+                        logger.error(f"[Subscription Check] AI text generation call limit exceeded for user {user_id}: {current_total_llm_calls}/{ai_text_gen_limit} (provider: {display_provider_name})")
+                        result = (False, f"AI text generation call limit reached. Used {current_total_llm_calls} of {ai_text_gen_limit} total AI text generation calls this billing period.", {
+                            'current_calls': current_total_llm_calls,
+                            'limit': ai_text_gen_limit,
+                            'usage_percentage': (current_total_llm_calls / ai_text_gen_limit) * 100 if ai_text_gen_limit > 0 else 0,
+                            'provider': display_provider_name,  # Use display name for consistency
+                            'usage_info': {
+                                'provider': display_provider_name,  # Use display name for user-facing info
+                                'current_calls': current_total_llm_calls,
+                                'limit': ai_text_gen_limit,
+                                'type': 'ai_text_generation',
+                                'breakdown': {
+                                    'gemini': usage.gemini_calls or 0,
+                                    'openai': usage.openai_calls or 0,
+                                    'anthropic': usage.anthropic_calls or 0,
+                                    'mistral': usage.mistral_calls or 0  # DB field name (not display name)
+                                }
+                            }
+                        })
+                        self.pricing_service._limits_cache[cache_key] = {
+                            'result': result,
+                            'expires_at': now + timedelta(seconds=30)
+                        }
+                        return result
+                    else:
+                        logger.debug(f"[Subscription Check] AI text generation limit check passed for user {user_id}: {current_total_llm_calls}/{ai_text_gen_limit if ai_text_gen_limit > 0 else 'unlimited'} (provider: {display_provider_name})")
+                else:
+                    # For non-LLM providers, check provider-specific limit
+                    current_calls = getattr(usage, f"{provider_name}_calls", 0) or 0
+                    call_limit = limits['limits'].get(f"{provider_name}_calls", 0) or 0
+                    
+                    # Only enforce limit if limit > 0 (0 means unlimited for Enterprise)
+                    if call_limit > 0 and current_calls >= call_limit:
+                        logger.error(f"[Subscription Check] Call limit exceeded for user {user_id}, provider {display_provider_name}: {current_calls}/{call_limit}")
+                        result = (False, f"API call limit reached for {display_provider_name}. Used {current_calls} of {call_limit} calls this billing period.", {
+                            'current_calls': current_calls,
+                            'limit': call_limit,
+                            'usage_percentage': 100.0,
+                            'provider': display_provider_name  # Use display name for consistency
+                        })
+                        self.pricing_service._limits_cache[cache_key] = {
+                            'result': result,
+                            'expires_at': now + timedelta(seconds=30)
+                        }
+                        return result
+                    else:
+                        logger.debug(f"[Subscription Check] Call limit check passed for user {user_id}, provider {display_provider_name}: {current_calls}/{call_limit if call_limit > 0 else 'unlimited'}")
+            except Exception as e:
+                logger.error(f"Error checking call limits: {e}")
+                # Continue to next check
+            
+            # Check token limits for LLM providers with error handling
+            # NOTE: token_limit = 0 means UNLIMITED (Enterprise plans)
+            try:
+                if provider in [APIProvider.GEMINI, APIProvider.OPENAI, APIProvider.ANTHROPIC, APIProvider.MISTRAL]:
+                    current_tokens = getattr(usage, f"{provider_name}_tokens", 0) or 0
+                    token_limit = limits['limits'].get(f"{provider_name}_tokens", 0) or 0
+                    
+                    # Only enforce limit if limit > 0 (0 means unlimited for Enterprise)
+                    if token_limit > 0 and (current_tokens + tokens_requested) > token_limit:
+                        result = (False, f"Token limit would be exceeded for {display_provider_name}. Current: {current_tokens}, Requested: {tokens_requested}, Limit: {token_limit}", {
+                            'current_tokens': current_tokens,
+                            'requested_tokens': tokens_requested,
+                            'limit': token_limit,
+                            'usage_percentage': ((current_tokens + tokens_requested) / token_limit) * 100,
+                            'provider': display_provider_name,  # Use display name in error details
+                            'usage_info': {
+                                'provider': display_provider_name,
+                                'current_tokens': current_tokens,
+                                'requested_tokens': tokens_requested,
+                                'limit': token_limit,
+                                'type': 'tokens'
+                            }
+                        })
+                        self.pricing_service._limits_cache[cache_key] = {
+                            'result': result,
+                            'expires_at': now + timedelta(seconds=30)
+                        }
+                        return result
+            except Exception as e:
+                logger.error(f"Error checking token limits: {e}")
+                # Continue to next check
+            
+            # Check cost limits with error handling
+            # NOTE: cost_limit = 0 means UNLIMITED (Enterprise plans)
+            try:
+                cost_limit = limits['limits'].get('monthly_cost', 0) or 0
+                # Only enforce limit if limit > 0 (0 means unlimited for Enterprise)
+                if cost_limit > 0 and usage.total_cost >= cost_limit:
+                    result = (False, f"Monthly cost limit reached. Current cost: ${usage.total_cost:.2f}, Limit: ${cost_limit:.2f}", {
+                        'current_cost': usage.total_cost,
+                        'limit': cost_limit,
+                        'usage_percentage': 100.0
+                    })
+                    self.pricing_service._limits_cache[cache_key] = {
+                        'result': result,
+                        'expires_at': now + timedelta(seconds=30)
+                    }
+                    return result
+            except Exception as e:
+                logger.error(f"Error checking cost limits: {e}")
+                # Continue to success case
+            
+            # Calculate usage percentages for warnings
+            try:
+                # Determine which call variables to use based on provider type
+                if is_llm_provider:
+                    # Use unified LLM call tracking
+                    current_call_count = current_total_llm_calls
+                    call_limit_value = ai_text_gen_limit
+                else:
+                    # Use provider-specific call tracking
+                    current_call_count = current_calls
+                    call_limit_value = call_limit
+                
+                call_usage_pct = (current_call_count / max(call_limit_value, 1)) * 100 if call_limit_value > 0 else 0
+                cost_usage_pct = (usage.total_cost / max(cost_limit, 1)) * 100 if cost_limit > 0 else 0
+                result = (True, "Within limits", {
+                    'current_calls': current_call_count,
+                    'call_limit': call_limit_value,
+                    'call_usage_percentage': call_usage_pct,
+                    'current_cost': usage.total_cost,
+                    'cost_limit': cost_limit,
+                    'cost_usage_percentage': cost_usage_pct
+                })
+                self.pricing_service._limits_cache[cache_key] = {
+                    'result': result,
+                    'expires_at': now + timedelta(seconds=30)
+                }
+                return result
+            except Exception as e:
+                logger.error(f"Error calculating usage percentages: {e}")
+                # Return basic success
+                return True, "Within limits", {}
+        
+        except Exception as e:
+            logger.error(f"Unexpected error in check_usage_limits for {user_id}: {e}")
+            # STRICT: Fail closed - deny requests if subscription system fails
+            return False, f"Subscription check error: {str(e)}", {}
+    
+    def check_comprehensive_limits(
+        self, 
+        user_id: str, 
+        operations: List[Dict[str, Any]]
+    ) -> Tuple[bool, Optional[str], Optional[Dict[str, Any]]]:
+        """
+        Comprehensive pre-flight validation that checks ALL limits before making ANY API calls.
+        
+        This prevents wasteful API calls by validating that ALL subsequent operations will succeed
+        before making the first external API call.
+        
+        Args:
+            user_id: User ID
+            operations: List of operations to validate, each with:
+                - 'provider': APIProvider enum
+                - 'tokens_requested': int (estimated tokens for LLM calls, 0 for non-LLM)
+                - 'actual_provider_name': Optional[str] (e.g., "huggingface" when provider is MISTRAL)
+                - 'operation_type': str (e.g., "google_grounding", "llm_call", "image_generation")
+        
+        Returns:
+            (can_proceed, error_message, error_details)
+            If can_proceed is False, error_message explains which limit would be exceeded
+        """
+        try:
+            logger.info(f"[Pre-flight Check] 🔍 Starting comprehensive validation for user {user_id}")
+            logger.info(f"[Pre-flight Check] 📋 Validating {len(operations)} operation(s) before making any API calls")
+            
+            # Get current usage and limits once
+            current_period = self.pricing_service.get_current_billing_period(user_id) or datetime.now().strftime("%Y-%m")
+            
+            logger.info(f"[Pre-flight Check] 📅 Billing Period: {current_period} (for user {user_id})")
+            
+            # Explicitly expire any cached objects and refresh from DB to ensure fresh data
+            self.db.expire_all()
+            
+            usage = self.db.query(UsageSummary).filter(
+                UsageSummary.user_id == user_id,
+                UsageSummary.billing_period == current_period
+            ).first()
+            
+            # CRITICAL: Explicitly refresh from database to get latest values (clears SQLAlchemy cache)
+            if usage:
+                self.db.refresh(usage)
+            
+            # Log what we actually read from database
+            if usage:
+                logger.info(f"[Pre-flight Check] 📊 Usage Summary from DB (Period: {current_period}):")
+                logger.info(f"   ├─ Gemini: {usage.gemini_tokens or 0} tokens / {usage.gemini_calls or 0} calls")
+                logger.info(f"   ├─ Mistral/HF: {usage.mistral_tokens or 0} tokens / {usage.mistral_calls or 0} calls")
+                logger.info(f"   ├─ Total Tokens: {usage.total_tokens or 0}")
+                logger.info(f"   └─ Usage Status: {usage.usage_status.value if usage.usage_status else 'N/A'}")
+            else:
+                logger.info(f"[Pre-flight Check] 📊 No usage summary found for period {current_period} (will create new)")
+            
+            if not usage:
+                # First usage this period, create summary
+                try:
+                    usage = UsageSummary(
+                        user_id=user_id,
+                        billing_period=current_period
+                    )
+                    self.db.add(usage)
+                    self.db.commit()
+                except Exception as create_error:
+                    logger.error(f"Error creating usage summary: {create_error}")
+                    self.db.rollback()
+                    return False, f"Failed to create usage summary: {str(create_error)}", {}
+            
+            # Get user limits
+            limits_dict = self.pricing_service.get_user_limits(user_id)
+            if not limits_dict:
+                # No subscription found - check for free tier
+                free_plan = self.db.query(SubscriptionPlan).filter(
+                    SubscriptionPlan.tier == SubscriptionTier.FREE,
+                    SubscriptionPlan.is_active == True
+                ).first()
+                if free_plan:
+                    limits_dict = self.pricing_service._plan_to_limits_dict(free_plan)
+                else:
+                    return False, "No subscription plan found. Please subscribe to a plan.", {}
+            
+            limits = limits_dict.get('limits', {})
+            
+            # Track cumulative usage across all operations
+            total_llm_calls = (
+                (usage.gemini_calls or 0) +
+                (usage.openai_calls or 0) +
+                (usage.anthropic_calls or 0) +
+                (usage.mistral_calls or 0)
+            )
+            total_llm_tokens = {}
+            total_images = usage.stability_calls or 0
+            
+            # Log current usage summary
+            logger.info(f"[Pre-flight Check] 📊 Current Usage Summary:")
+            logger.info(f"   └─ Total LLM Calls: {total_llm_calls}")
+            logger.info(f"   └─ Gemini Tokens: {usage.gemini_tokens or 0}, Mistral/HF Tokens: {usage.mistral_tokens or 0}")
+            logger.info(f"   └─ Image Calls: {total_images}")
+            
+            # Validate each operation
+            for op_idx, operation in enumerate(operations):
+                provider = operation.get('provider')
+                provider_name = provider.value if hasattr(provider, 'value') else str(provider)
+                tokens_requested = operation.get('tokens_requested', 0)
+                actual_provider_name = operation.get('actual_provider_name')
+                operation_type = operation.get('operation_type', 'unknown')
+                
+                display_provider_name = actual_provider_name or provider_name
+                
+                logger.error(f"[Pre-flight Check] ✅ Operation {op_idx + 1}/{len(operations)}: {operation_type}")
+                logger.error(f"   ├─ Provider: {display_provider_name} (enum: {provider_name})")
+                logger.error(f"   ├─ Operation Index: {op_idx}")
+                logger.error(f"   └─ Estimated Tokens Requested: {tokens_requested}")
+                
+                # Check if this is an LLM provider
+                llm_providers = ['gemini', 'openai', 'anthropic', 'mistral']
+                is_llm_provider = provider_name in llm_providers
+                
+                # Check unified AI text generation limit for LLM providers
+                if is_llm_provider:
+                    ai_text_gen_limit = limits.get('ai_text_generation_calls', 0) or 0
+                    if ai_text_gen_limit == 0:
+                        # Fallback to provider-specific limit
+                        ai_text_gen_limit = limits.get(f"{provider_name}_calls", 0) or 0
+                    
+                    # Count this operation as an LLM call
+                    projected_total_llm_calls = total_llm_calls + 1
+                    
+                    if ai_text_gen_limit > 0 and projected_total_llm_calls > ai_text_gen_limit:
+                        error_info = {
+                            'current_calls': total_llm_calls,
+                            'limit': ai_text_gen_limit,
+                            'provider': display_provider_name,
+                            'operation_type': operation_type,
+                            'operation_index': op_idx
+                        }
+                        return False, f"AI text generation call limit would be exceeded. Would use {projected_total_llm_calls} of {ai_text_gen_limit} total AI text generation calls.", {
+                            'error_type': 'call_limit',
+                            'usage_info': error_info
+                        }
+                    
+                    # Check token limits for this provider
+                    # CRITICAL: Always query fresh from DB for each operation to avoid SQLAlchemy cache issues
+                    # This ensures we get the latest values after subscription renewal, even for cumulative tracking
+                    provider_tokens_key = f"{provider_name}_tokens"
+                    
+                    # Try to get fresh value from DB with comprehensive error handling
+                    base_current_tokens = 0
+                    query_succeeded = False
+                    
+                    try:
+                        # Validate column name is safe (only allow known provider token columns)
+                        valid_token_columns = ['gemini_tokens', 'openai_tokens', 'anthropic_tokens', 'mistral_tokens']
+                        
+                        if provider_tokens_key not in valid_token_columns:
+                            logger.error(f"   └─ Invalid provider tokens key: {provider_tokens_key}")
+                            query_succeeded = True  # Treat as success with 0 value
+                        else:
+                            # Method 1: Try raw SQL query to completely bypass ORM cache
+                            try:
+                                logger.debug(f"   └─ Attempting raw SQL query for {provider_tokens_key}")
+                                sql_query = text(f"""
+                                    SELECT {provider_tokens_key} 
+                                    FROM usage_summaries 
+                                    WHERE user_id = :user_id 
+                                    AND billing_period = :period
+                                    LIMIT 1
+                                """)
+                                
+                                logger.debug(f"   └─ SQL: SELECT {provider_tokens_key} FROM usage_summaries WHERE user_id={user_id} AND billing_period={current_period}")
+                                
+                                result = self.db.execute(sql_query, {
+                                    'user_id': user_id,
+                                    'period': current_period
+                                }).first()
+                                
+                                if result:
+                                    base_current_tokens = result[0] if result[0] is not None else 0
+                                    logger.error(f"[Pre-flight Check] ✅ Raw SQL query returned result: {result[0]} -> {base_current_tokens}")
+                                else:
+                                    base_current_tokens = 0
+                                    logger.error(f"[Pre-flight Check] ⚠️  Raw SQL query returned None (no rows found)")
+                                
+                                query_succeeded = True
+                                logger.error(f"[Pre-flight Check] ✅ Raw SQL query succeeded for {provider_tokens_key}: {base_current_tokens}")
+                                
+                            except Exception as sql_error:
+                                logger.error(f"   └─ Raw SQL query failed for {provider_tokens_key}: {type(sql_error).__name__}: {sql_error}", exc_info=True)
+                                query_succeeded = False  # Will try ORM fallback
+                            
+                            # Method 2: Fallback to fresh ORM query if raw SQL fails
+                            if not query_succeeded:
+                                try:
+                                    # Expire all cached objects and do fresh query
+                                    self.db.expire_all()
+                                    fresh_usage = self.db.query(UsageSummary).filter(
+                                        UsageSummary.user_id == user_id,
+                                        UsageSummary.billing_period == current_period
+                                    ).first()
+                                    
+                                    if fresh_usage:
+                                        # Explicitly refresh to get latest from DB
+                                        self.db.refresh(fresh_usage)
+                                        base_current_tokens = getattr(fresh_usage, provider_tokens_key, 0) or 0
+                                    else:
+                                        base_current_tokens = 0
+                                    
+                                    query_succeeded = True
+                                    logger.info(f"[Pre-flight Check] ✅ ORM fallback query succeeded for {provider_tokens_key}: {base_current_tokens}")
+                                    
+                                except Exception as orm_error:
+                                    logger.error(f"   └─ ORM query also failed: {orm_error}", exc_info=True)
+                                    query_succeeded = False
+                    
+                    except Exception as e:
+                        logger.error(f"   └─ Unexpected error getting tokens from DB for {provider_tokens_key}: {e}", exc_info=True)
+                        base_current_tokens = 0  # Fail safe - assume 0 if we can't query
+                    
+                    if not query_succeeded:
+                        logger.warning(f"   └─ Both query methods failed, using 0 as fallback")
+                    
+                    # CRITICAL LOG: Always log what we got from DB - this helps debug renewal issues
+                    # Use ERROR level to ensure it shows even if INFO is filtered
+                    logger.error(f"[Pre-flight Check] 🔍 Fresh DB Query for {display_provider_name}:")
+                    logger.error(f"   ├─ Column: {provider_tokens_key}")
+                    logger.error(f"   ├─ Billing Period: {current_period}")
+                    logger.error(f"   ├─ User ID: {user_id}")
+                    logger.error(f"   ├─ Method: {'Raw SQL' if query_succeeded and base_current_tokens >= 0 else 'ORM' if query_succeeded else 'Failed - using 0'}")
+                    logger.error(f"   └─ Value from DB: {base_current_tokens}")
+                    
+                    # Add any projected tokens from previous operations in this validation run
+                    # Note: total_llm_tokens tracks ONLY projected tokens from this run, not base DB value
+                    projected_from_previous = total_llm_tokens.get(provider_tokens_key, 0)
+                    
+                    # Current tokens = base from DB + projected from previous operations in this run
+                    current_provider_tokens = base_current_tokens + projected_from_previous
+                    
+                    # Use ERROR level to ensure visibility
+                    logger.error(f"[Pre-flight Check] 📊 Token Calculation for {display_provider_name}:")
+                    logger.error(f"   ├─ Base from DB (fresh query): {base_current_tokens}")
+                    logger.error(f"   ├─ Projected from previous ops in this run: {projected_from_previous}")
+                    logger.error(f"   └─ Total current tokens (base + projected): {current_provider_tokens}")
+                    
+                    # Also check the initial usage object to see if it's being used incorrectly
+                    if usage and hasattr(usage, provider_tokens_key):
+                        initial_usage_value = getattr(usage, provider_tokens_key, 0) or 0
+                        logger.error(f"   ⚠️  Initial usage object value: {initial_usage_value} (this should NOT be used for fresh query)")
+                    
+                    token_limit = limits.get(provider_tokens_key, 0) or 0
+                    
+                    if token_limit > 0 and tokens_requested > 0:
+                        projected_tokens = current_provider_tokens + tokens_requested
+                        logger.info(f"   └─ Token Check: {current_provider_tokens} (current) + {tokens_requested} (requested) = {projected_tokens} (total) / {token_limit} (limit)")
+                        
+                        if projected_tokens > token_limit:
+                            usage_percentage = (projected_tokens / token_limit) * 100 if token_limit > 0 else 0
+                            error_info = {
+                                'current_tokens': current_provider_tokens,
+                                'base_tokens_from_db': base_current_tokens,
+                                'projected_from_previous_ops': projected_from_previous,
+                                'requested_tokens': tokens_requested,
+                                'limit': token_limit,
+                                'provider': display_provider_name,
+                                'operation_type': operation_type,
+                                'operation_index': op_idx
+                            }
+                            # Make error message clearer: show actual DB usage vs projected
+                            if projected_from_previous > 0:
+                                error_msg = (
+                                    f"Token limit exceeded for {display_provider_name} "
+                                    f"({operation_type}). "
+                                    f"Base usage: {base_current_tokens}/{token_limit}, "
+                                    f"After previous operations in this workflow: {current_provider_tokens}/{token_limit}, "
+                                    f"This operation would add: {tokens_requested}, "
+                                    f"Total would be: {projected_tokens} (exceeds by {projected_tokens - token_limit} tokens)"
+                                )
+                            else:
+                                error_msg = (
+                                    f"Token limit exceeded for {display_provider_name} "
+                                    f"({operation_type}). "
+                                    f"Current: {current_provider_tokens}/{token_limit}, "
+                                    f"Requested: {tokens_requested}, "
+                                    f"Would exceed by: {projected_tokens - token_limit} tokens "
+                                    f"({usage_percentage:.1f}% of limit)"
+                                )
+                            logger.error(f"[Pre-flight Check] ❌ BLOCKED: {error_msg}")
+                            return False, error_msg, {
+                                'error_type': 'token_limit',
+                                'usage_info': error_info
+                            }
+                        else:
+                            logger.info(f"   └─ ✅ Token limit check passed: {projected_tokens} <= {token_limit}")
+                    
+                    # Update cumulative counts for next operation
+                    total_llm_calls = projected_total_llm_calls
+                    # Update cumulative projected tokens from this validation run
+                    # This represents only projected tokens from previous operations in this run
+                    # Base DB value is always queried fresh, so we only track the projection delta
+                    old_projected = total_llm_tokens.get(provider_tokens_key, 0)
+                    if tokens_requested > 0:
+                        # Add this operation's tokens to cumulative projected tokens
+                        total_llm_tokens[provider_tokens_key] = projected_from_previous + tokens_requested
+                        logger.error(f"[Pre-flight Check] 📝 Updated cumulative projected tokens for {display_provider_name}:")
+                        logger.error(f"   ├─ Previous projected: {projected_from_previous}")
+                        logger.error(f"   ├─ This operation requested: {tokens_requested}")
+                        logger.error(f"   ├─ New cumulative projected: {total_llm_tokens[provider_tokens_key]}")
+                        logger.error(f"   └─ Old value in dict was: {old_projected}")
+                    else:
+                        # No tokens requested, keep existing projected tokens (or 0 if first operation)
+                        total_llm_tokens[provider_tokens_key] = projected_from_previous
+                        logger.error(f"[Pre-flight Check] 📝 No tokens requested, keeping projected at: {projected_from_previous}")
+                
+                # Check image generation limits
+                elif provider == APIProvider.STABILITY:
+                    image_limit = limits.get('stability_calls', 0) or 0
+                    projected_images = total_images + 1
+                    
+                    if image_limit > 0 and projected_images > image_limit:
+                        error_info = {
+                            'current_images': total_images,
+                            'limit': image_limit,
+                            'provider': 'stability',
+                            'operation_type': operation_type,
+                            'operation_index': op_idx
+                        }
+                        return False, f"Image generation limit would be exceeded. Would use {projected_images} of {image_limit} images this billing period.", {
+                            'error_type': 'image_limit',
+                            'usage_info': error_info
+                        }
+                    
+                    total_images = projected_images
+                
+                # Check other provider-specific limits
+                else:
+                    provider_calls_key = f"{provider_name}_calls"
+                    current_provider_calls = getattr(usage, provider_calls_key, 0) or 0
+                    call_limit = limits.get(provider_calls_key, 0) or 0
+                    
+                    if call_limit > 0:
+                        projected_calls = current_provider_calls + 1
+                        if projected_calls > call_limit:
+                            error_info = {
+                                'current_calls': current_provider_calls,
+                                'limit': call_limit,
+                                'provider': display_provider_name,
+                                'operation_type': operation_type,
+                                'operation_index': op_idx
+                            }
+                            return False, f"API call limit would be exceeded for {display_provider_name}. Would use {projected_calls} of {call_limit} calls this billing period.", {
+                                'error_type': 'call_limit',
+                                'usage_info': error_info
+                            }
+            
+            # All checks passed
+            logger.info(f"[Pre-flight Check] ✅ All {len(operations)} operation(s) validated successfully")
+            logger.info(f"[Pre-flight Check] ✅ User {user_id} is cleared to proceed with API calls")
+            return True, None, None
+            
+        except Exception as e:
+            error_type = type(e).__name__
+            error_message = str(e)
+            logger.error(f"[Pre-flight Check] ❌ Error during comprehensive limit check: {error_type}: {error_message}", exc_info=True)
+            logger.error(f"[Pre-flight Check] ❌ User: {user_id}, Operations count: {len(operations) if operations else 0}")
+            return False, f"Failed to validate limits: {error_type}: {error_message}", {}
+
diff --git a/backend/services/subscription/preflight_validator.py b/backend/services/subscription/preflight_validator.py
index 0526cc3c..96f24f7d 100644
--- a/backend/services/subscription/preflight_validator.py
+++ b/backend/services/subscription/preflight_validator.py
@@ -44,15 +44,17 @@ def validate_research_operations(
             llm_provider_name = "gemini"
         
         # Estimate tokens for each operation in research workflow
-        # Google Grounding call: ~2000 tokens (input + output)
+        # Google Grounding call: ~1200 tokens (input: ~500 tokens, output: ~700 tokens for research results)
         # Keyword analyzer: ~1000 tokens (input: 3000 chars research, output: structured JSON)
         # Competitor analyzer: ~1000 tokens (input: 3000 chars research, output: structured JSON)
         # Content angle generator: ~1000 tokens (input: 3000 chars research, output: list of angles)
+        # Note: These are conservative estimates. Actual usage may be lower, but we use these for pre-flight validation
+        # to prevent wasteful API calls if the workflow would exceed limits.
         
         operations_to_validate = [
             {
                 'provider': APIProvider.GEMINI,  # Google Grounding uses Gemini
-                'tokens_requested': 2000,
+                'tokens_requested': 1200,  # Reduced from 2000 to more realistic estimate
                 'actual_provider_name': 'gemini',
                 'operation_type': 'google_grounding'
             },
@@ -126,6 +128,120 @@ def validate_research_operations(
         )
 
 
+def validate_exa_research_operations(
+    pricing_service: PricingService,
+    user_id: str,
+    gpt_provider: str = "google"
+) -> None:
+    """
+    Validate all operations for an Exa research workflow before making ANY API calls.
+    
+    This prevents wasteful external API calls (like Exa search) if subsequent
+    LLM calls would fail due to token or call limits.
+    
+    Args:
+        pricing_service: PricingService instance
+        user_id: User ID for subscription checking
+        gpt_provider: GPT provider from env var (defaults to "google")
+        
+    Returns:
+        None
+        If validation fails, raises HTTPException with 429 status
+    """
+    try:
+        # Determine actual provider for LLM calls based on GPT_PROVIDER env var
+        gpt_provider_lower = gpt_provider.lower()
+        if gpt_provider_lower == "huggingface":
+            llm_provider_enum = APIProvider.MISTRAL  # Maps to HuggingFace
+            llm_provider_name = "huggingface"
+        else:
+            llm_provider_enum = APIProvider.GEMINI
+            llm_provider_name = "gemini"
+        
+        # Estimate tokens for each operation in Exa research workflow
+        # Exa Search call: 1 Exa API call (not token-based)
+        # Keyword analyzer: ~1000 tokens (input: research results, output: structured JSON)
+        # Competitor analyzer: ~1000 tokens (input: research results, output: structured JSON)
+        # Content angle generator: ~1000 tokens (input: research results, output: list of angles)
+        # Note: These are conservative estimates for pre-flight validation
+        
+        operations_to_validate = [
+            {
+                'provider': APIProvider.EXA,  # Exa API call
+                'tokens_requested': 0,
+                'actual_provider_name': 'exa',
+                'operation_type': 'exa_neural_search'
+            },
+            {
+                'provider': llm_provider_enum,
+                'tokens_requested': 1000,
+                'actual_provider_name': llm_provider_name,
+                'operation_type': 'keyword_analysis'
+            },
+            {
+                'provider': llm_provider_enum,
+                'tokens_requested': 1000,
+                'actual_provider_name': llm_provider_name,
+                'operation_type': 'competitor_analysis'
+            },
+            {
+                'provider': llm_provider_enum,
+                'tokens_requested': 1000,
+                'actual_provider_name': llm_provider_name,
+                'operation_type': 'content_angle_generation'
+            }
+        ]
+        
+        logger.info(f"[Pre-flight Validator] 🚀 Starting Exa Research Workflow Validation")
+        logger.info(f"   ├─ User: {user_id}")
+        logger.info(f"   ├─ LLM Provider: {llm_provider_name} (GPT_PROVIDER={gpt_provider})")
+        logger.info(f"   └─ Operations to validate: {len(operations_to_validate)}")
+        
+        can_proceed, message, error_details = pricing_service.check_comprehensive_limits(
+            user_id=user_id,
+            operations=operations_to_validate
+        )
+        
+        if not can_proceed:
+            usage_info = error_details.get('usage_info', {}) if error_details else {}
+            provider = usage_info.get('provider', llm_provider_name) if usage_info else llm_provider_name
+            operation_type = usage_info.get('operation_type', 'unknown')
+            
+            logger.error(f"[Pre-flight Validator] ❌ EXA RESEARCH WORKFLOW BLOCKED")
+            logger.error(f"   ├─ User: {user_id}")
+            logger.error(f"   ├─ Blocked at: {operation_type}")
+            logger.error(f"   ├─ Provider: {provider}")
+            logger.error(f"   └─ Reason: {message}")
+            
+            # Raise HTTPException immediately - frontend gets immediate response, no API calls made
+            raise HTTPException(
+                status_code=429,
+                detail={
+                    'error': message,
+                    'message': message,
+                    'provider': provider,
+                    'usage_info': usage_info if usage_info else error_details
+                }
+            )
+        
+        logger.info(f"[Pre-flight Validator] ✅ EXA RESEARCH WORKFLOW APPROVED")
+        logger.info(f"   ├─ User: {user_id}")
+        logger.info(f"   └─ All {len(operations_to_validate)} operations validated - proceeding with API calls")
+        # Validation passed - no return needed (function raises HTTPException if validation fails)
+        
+    except HTTPException:
+        raise
+    except Exception as e:
+        logger.error(f"[Pre-flight Validator] Error validating Exa research operations: {e}", exc_info=True)
+        raise HTTPException(
+            status_code=500,
+            detail={
+                'error': f"Failed to validate operations: {str(e)}",
+                'message': f"Failed to validate operations: {str(e)}"
+            }
+        )
+
+
 def validate_image_generation_operations(
     pricing_service: PricingService,
     user_id: str
diff --git a/backend/services/subscription/pricing_service.py b/backend/services/subscription/pricing_service.py
index 629759d5..a72912c5 100644
--- a/backend/services/subscription/pricing_service.py
+++ b/backend/services/subscription/pricing_service.py
@@ -258,6 +258,12 @@ class PricingService:
                 "model_name": "stable-diffusion",
                 "cost_per_image": 0.04,  # $0.04 per image
                 "description": "Stability AI Image Generation"
+            },
+            {
+                "provider": APIProvider.EXA,
+                "model_name": "exa-search",
+                "cost_per_request": 0.005,  # $0.005 per search (1-25 results)
+                "description": "Exa Neural Search API"
             }
         ]
         
@@ -296,6 +302,7 @@ class PricingService:
                 "metaphor_calls_limit": 10,
                 "firecrawl_calls_limit": 10,
                 "stability_calls_limit": 5,
+                "exa_calls_limit": 100,
                 "gemini_tokens_limit": 100000,
                 "monthly_cost_limit": 0.0,
                 "features": ["basic_content_generation", "limited_research"],
@@ -316,10 +323,11 @@ class PricingService:
                 "metaphor_calls_limit": 100,
                 "firecrawl_calls_limit": 100,
                 "stability_calls_limit": 5,
-                "gemini_tokens_limit": 2000,
-                "openai_tokens_limit": 2000,
-                "anthropic_tokens_limit": 2000,
-                "mistral_tokens_limit": 2000,
+                "exa_calls_limit": 500,
+                "gemini_tokens_limit": 20000,  # Increased from 5000 for better stability
+                "openai_tokens_limit": 20000,  # Increased from 5000 for better stability
+                "anthropic_tokens_limit": 20000,  # Increased from 5000 for better stability
+                "mistral_tokens_limit": 20000,  # Increased from 5000 for better stability
                 "monthly_cost_limit": 50.0,
                 "features": ["full_content_generation", "advanced_research", "basic_analytics"],
                 "description": "Great for individuals and small teams"
@@ -338,6 +346,7 @@ class PricingService:
                 "metaphor_calls_limit": 500,
                 "firecrawl_calls_limit": 500,
                 "stability_calls_limit": 200,
+                "exa_calls_limit": 2000,
                 "gemini_tokens_limit": 5000000,
                 "openai_tokens_limit": 2500000,
                 "anthropic_tokens_limit": 1000000,
@@ -360,6 +369,7 @@ class PricingService:
                 "metaphor_calls_limit": 0,
                 "firecrawl_calls_limit": 0,
                 "stability_calls_limit": 0,
+                "exa_calls_limit": 0,  # Unlimited
                 "gemini_tokens_limit": 0,
                 "openai_tokens_limit": 0,
                 "anthropic_tokens_limit": 0,
@@ -423,11 +433,14 @@ class PricingService:
     def get_user_limits(self, user_id: str) -> Optional[Dict[str, Any]]:
         """Get usage limits for a user based on their subscription."""
         
+        # CRITICAL: Expire all objects first to ensure fresh data after renewal
+        self.db.expire_all()
+        
         subscription = self.db.query(UserSubscription).filter(
             UserSubscription.user_id == user_id,
             UserSubscription.is_active == True
         ).first()
-        
+
         if not subscription:
             # Return free tier limits
             free_plan = self.db.query(SubscriptionPlan).filter(
@@ -439,7 +452,23 @@ class PricingService:
 
         # Ensure current period before returning limits
         self._ensure_subscription_current(subscription)
-        return self._plan_to_limits_dict(subscription.plan)
+        
+        # CRITICAL: Refresh subscription to get latest plan_id, then refresh plan relationship
+        self.db.refresh(subscription)
+        
+        # Re-query plan directly to ensure fresh data (bypass relationship cache)
+        plan = self.db.query(SubscriptionPlan).filter(
+            SubscriptionPlan.id == subscription.plan_id
+        ).first()
+        
+        if not plan:
+            logger.error(f"Plan not found for subscription plan_id={subscription.plan_id}")
+            return None
+        
+        # Refresh plan to ensure fresh limits
+        self.db.refresh(plan)
+        
+        return self._plan_to_limits_dict(plan)
     
     def _ensure_ai_text_gen_column_detection(self) -> None:
         """Detect at runtime whether ai_text_generation_calls_limit column exists and cache the result."""
@@ -508,290 +537,20 @@ class PricingService:
                           tokens_requested: int = 0, actual_provider_name: Optional[str] = None) -> Tuple[bool, str, Dict[str, Any]]:
         """Check if user can make an API call within their limits.
         
+        Delegates to LimitValidator for actual validation logic.
+        
         Args:
             user_id: User ID
             provider: APIProvider enum (may be MISTRAL for HuggingFace)
             tokens_requested: Estimated tokens for the request
             actual_provider_name: Optional actual provider name (e.g., "huggingface" when provider is MISTRAL)
-        """
-        try:
-            # Use actual_provider_name if provided, otherwise use enum value
-            # This fixes cases where HuggingFace maps to MISTRAL enum but should show as "huggingface" in errors
-            display_provider_name = actual_provider_name or provider.value
-            
-            logger.debug(f"[Subscription Check] Starting limit check for user {user_id}, provider {display_provider_name}, tokens {tokens_requested}")
-            
-            # Short TTL cache to reduce DB reads under sustained traffic
-            cache_key = f"{user_id}:{provider.value}"
-            now = datetime.utcnow()
-            cached = self._limits_cache.get(cache_key)
-            if cached and cached.get('expires_at') and cached['expires_at'] > now:
-                logger.debug(f"[Subscription Check] Using cached result for {user_id}:{provider.value}")
-                return tuple(cached['result'])  # type: ignore
-
-            # Get user subscription first to check expiration
-            subscription = self.db.query(UserSubscription).filter(
-                UserSubscription.user_id == user_id,
-                UserSubscription.is_active == True
-            ).first()
-            
-            if subscription:
-                logger.debug(f"[Subscription Check] Found subscription for user {user_id}: plan_id={subscription.plan_id}, period_end={subscription.current_period_end}")
-            else:
-                logger.debug(f"[Subscription Check] No active subscription found for user {user_id}")
-            
-            # Check subscription expiration (STRICT: deny if expired)
-            if subscription:
-                if subscription.current_period_end < now:
-                    logger.warning(f"[Subscription Check] Subscription expired for user {user_id}: period_end={subscription.current_period_end}, now={now}")
-                    # Subscription expired - check if auto_renew is enabled
-                    if not getattr(subscription, 'auto_renew', False):
-                        # Expired and no auto-renew - deny access
-                        logger.warning(f"[Subscription Check] Subscription expired for user {user_id}, auto_renew=False, denying access")
-                        result = (False, "Subscription expired. Please renew your subscription to continue using the service.", {
-                            'expired': True,
-                            'period_end': subscription.current_period_end.isoformat()
-                        })
-                        self._limits_cache[cache_key] = {
-                            'result': result,
-                            'expires_at': now + timedelta(seconds=30)
-                        }
-                        return result
-                    else:
-                        # Try to auto-renew
-                        if not self._ensure_subscription_current(subscription):
-                            # Auto-renew failed - deny access
-                            result = (False, "Subscription expired and auto-renewal failed. Please renew manually.", {
-                                'expired': True,
-                                'auto_renew_failed': True
-                            })
-                            self._limits_cache[cache_key] = {
-                                'result': result,
-                                'expires_at': now + timedelta(seconds=30)
-                            }
-                            return result
-
-            # Get user limits with error handling (STRICT: fail on errors)
-            try:
-                limits = self.get_user_limits(user_id)
-                if limits:
-                    logger.debug(f"[Subscription Check] Retrieved limits for user {user_id}: plan={limits.get('plan_name')}, tier={limits.get('tier')}")
-                else:
-                    logger.debug(f"[Subscription Check] No limits found for user {user_id}, checking free tier")
-            except Exception as e:
-                logger.error(f"[Subscription Check] Error getting user limits for {user_id}: {e}", exc_info=True)
-                # STRICT: Fail closed - deny request if we can't check limits
-                return False, f"Failed to retrieve subscription limits: {str(e)}", {}
-            
-            if not limits:
-                # No subscription found - check for free tier
-                free_plan = self.db.query(SubscriptionPlan).filter(
-                    SubscriptionPlan.tier == SubscriptionTier.FREE,
-                    SubscriptionPlan.is_active == True
-                ).first()
-                if free_plan:
-                    logger.info(f"[Subscription Check] Assigning free tier to user {user_id}")
-                    limits = self._plan_to_limits_dict(free_plan)
-                else:
-                    # No subscription and no free tier - deny access
-                    logger.warning(f"[Subscription Check] No subscription or free tier found for user {user_id}, denying access")
-                    return False, "No subscription plan found. Please subscribe to a plan.", {}
-            
-            # Get current usage for this billing period with error handling
-            try:
-                current_period = self.get_current_billing_period(user_id) or datetime.now().strftime("%Y-%m")
-                usage = self.db.query(UsageSummary).filter(
-                    UsageSummary.user_id == user_id,
-                    UsageSummary.billing_period == current_period
-                ).first()
-                
-                if not usage:
-                    # First usage this period, create summary
-                    try:
-                        usage = UsageSummary(
-                            user_id=user_id,
-                            billing_period=current_period
-                        )
-                        self.db.add(usage)
-                        self.db.commit()
-                    except Exception as create_error:
-                        logger.error(f"Error creating usage summary: {create_error}")
-                        self.db.rollback()
-                        # STRICT: Fail closed on DB error
-                        return False, f"Failed to create usage summary: {str(create_error)}", {}
-            except Exception as e:
-                logger.error(f"Error getting usage summary for {user_id}: {e}")
-                self.db.rollback()
-                # STRICT: Fail closed on DB error
-                return False, f"Failed to retrieve usage summary: {str(e)}", {}
-            
-            # Check call limits with error handling
-            # NOTE: call_limit = 0 means UNLIMITED (Enterprise plans)
-            try:
-                # Use display_provider_name for error messages, but provider.value for DB queries
-                provider_name = provider.value  # For DB field names (e.g., "mistral_calls", "mistral_tokens")
-                
-                # For LLM text generation providers, check against unified total_calls limit
-                llm_providers = ['gemini', 'openai', 'anthropic', 'mistral']
-                is_llm_provider = provider_name in llm_providers
-                
-                if is_llm_provider:
-                    # Use unified AI text generation limit (total_calls across all LLM providers)
-                    ai_text_gen_limit = limits['limits'].get('ai_text_generation_calls', 0) or 0
-                    
-                    # If unified limit not set, fall back to provider-specific limit for backwards compatibility
-                    if ai_text_gen_limit == 0:
-                        ai_text_gen_limit = limits['limits'].get(f"{provider_name}_calls", 0) or 0
-                    
-                    # Calculate total LLM provider calls (sum of gemini + openai + anthropic + mistral)
-                    current_total_llm_calls = (
-                        (usage.gemini_calls or 0) +
-                        (usage.openai_calls or 0) +
-                        (usage.anthropic_calls or 0) +
-                        (usage.mistral_calls or 0)
-                    )
-                    
-                    # Only enforce limit if limit > 0 (0 means unlimited for Enterprise)
-                    if ai_text_gen_limit > 0 and current_total_llm_calls >= ai_text_gen_limit:
-                        logger.error(f"[Subscription Check] AI text generation call limit exceeded for user {user_id}: {current_total_llm_calls}/{ai_text_gen_limit} (provider: {display_provider_name})")
-                        result = (False, f"AI text generation call limit reached. Used {current_total_llm_calls} of {ai_text_gen_limit} total AI text generation calls this billing period.", {
-                            'current_calls': current_total_llm_calls,
-                            'limit': ai_text_gen_limit,
-                            'usage_percentage': (current_total_llm_calls / ai_text_gen_limit) * 100 if ai_text_gen_limit > 0 else 0,
-                            'provider': display_provider_name,  # Use display name for consistency
-                            'usage_info': {
-                                'provider': display_provider_name,  # Use display name for user-facing info
-                                'current_calls': current_total_llm_calls,
-                                'limit': ai_text_gen_limit,
-                                'type': 'ai_text_generation',
-                                'breakdown': {
-                                    'gemini': usage.gemini_calls or 0,
-                                    'openai': usage.openai_calls or 0,
-                                    'anthropic': usage.anthropic_calls or 0,
-                                    'mistral': usage.mistral_calls or 0  # DB field name (not display name)
-                                }
-                            }
-                        })
-                        self._limits_cache[cache_key] = {
-                            'result': result,
-                            'expires_at': now + timedelta(seconds=30)
-                        }
-                        return result
-                    else:
-                        logger.debug(f"[Subscription Check] AI text generation limit check passed for user {user_id}: {current_total_llm_calls}/{ai_text_gen_limit if ai_text_gen_limit > 0 else 'unlimited'} (provider: {display_provider_name})")
-                else:
-                    # For non-LLM providers, check provider-specific limit
-                    current_calls = getattr(usage, f"{provider_name}_calls", 0) or 0
-                    call_limit = limits['limits'].get(f"{provider_name}_calls", 0) or 0
-                    
-                    # Only enforce limit if limit > 0 (0 means unlimited for Enterprise)
-                    if call_limit > 0 and current_calls >= call_limit:
-                        logger.error(f"[Subscription Check] Call limit exceeded for user {user_id}, provider {display_provider_name}: {current_calls}/{call_limit}")
-                        result = (False, f"API call limit reached for {display_provider_name}. Used {current_calls} of {call_limit} calls this billing period.", {
-                            'current_calls': current_calls,
-                            'limit': call_limit,
-                            'usage_percentage': 100.0,
-                            'provider': display_provider_name  # Use display name for consistency
-                        })
-                        self._limits_cache[cache_key] = {
-                            'result': result,
-                            'expires_at': now + timedelta(seconds=30)
-                        }
-                        return result
-                    else:
-                        logger.debug(f"[Subscription Check] Call limit check passed for user {user_id}, provider {display_provider_name}: {current_calls}/{call_limit if call_limit > 0 else 'unlimited'}")
-            except Exception as e:
-                logger.error(f"Error checking call limits: {e}")
-                # Continue to next check
-            
-            # Check token limits for LLM providers with error handling
-            # NOTE: token_limit = 0 means UNLIMITED (Enterprise plans)
-            try:
-                if provider in [APIProvider.GEMINI, APIProvider.OPENAI, APIProvider.ANTHROPIC, APIProvider.MISTRAL]:
-                    current_tokens = getattr(usage, f"{provider_name}_tokens", 0) or 0
-                    token_limit = limits['limits'].get(f"{provider_name}_tokens", 0) or 0
-                    
-                    # Only enforce limit if limit > 0 (0 means unlimited for Enterprise)
-                    if token_limit > 0 and (current_tokens + tokens_requested) > token_limit:
-                        result = (False, f"Token limit would be exceeded for {display_provider_name}. Current: {current_tokens}, Requested: {tokens_requested}, Limit: {token_limit}", {
-                            'current_tokens': current_tokens,
-                            'requested_tokens': tokens_requested,
-                            'limit': token_limit,
-                            'usage_percentage': ((current_tokens + tokens_requested) / token_limit) * 100,
-                            'provider': display_provider_name,  # Use display name in error details
-                            'usage_info': {
-                                'provider': display_provider_name,
-                                'current_tokens': current_tokens,
-                                'requested_tokens': tokens_requested,
-                                'limit': token_limit,
-                                'type': 'tokens'
-                            }
-                        })
-                        self._limits_cache[cache_key] = {
-                            'result': result,
-                            'expires_at': now + timedelta(seconds=30)
-                        }
-                        return result
-            except Exception as e:
-                logger.error(f"Error checking token limits: {e}")
-                # Continue to next check
-            
-            # Check cost limits with error handling
-            # NOTE: cost_limit = 0 means UNLIMITED (Enterprise plans)
-            try:
-                cost_limit = limits['limits'].get('monthly_cost', 0) or 0
-                # Only enforce limit if limit > 0 (0 means unlimited for Enterprise)
-                if cost_limit > 0 and usage.total_cost >= cost_limit:
-                    result = (False, f"Monthly cost limit reached. Current cost: ${usage.total_cost:.2f}, Limit: ${cost_limit:.2f}", {
-                        'current_cost': usage.total_cost,
-                        'limit': cost_limit,
-                        'usage_percentage': 100.0
-                    })
-                    self._limits_cache[cache_key] = {
-                        'result': result,
-                        'expires_at': now + timedelta(seconds=30)
-                    }
-                    return result
-            except Exception as e:
-                logger.error(f"Error checking cost limits: {e}")
-                # Continue to success case
-            
-            # Calculate usage percentages for warnings
-            try:
-                # Determine which call variables to use based on provider type
-                if is_llm_provider:
-                    # Use unified LLM call tracking
-                    current_call_count = current_total_llm_calls
-                    call_limit_value = ai_text_gen_limit
-                else:
-                    # Use provider-specific call tracking
-                    current_call_count = current_calls
-                    call_limit_value = call_limit
-                
-                call_usage_pct = (current_call_count / max(call_limit_value, 1)) * 100 if call_limit_value > 0 else 0
-                cost_usage_pct = (usage.total_cost / max(cost_limit, 1)) * 100 if cost_limit > 0 else 0
-                result = (True, "Within limits", {
-                    'current_calls': current_call_count,
-                    'call_limit': call_limit_value,
-                    'call_usage_percentage': call_usage_pct,
-                    'current_cost': usage.total_cost,
-                    'cost_limit': cost_limit,
-                    'cost_usage_percentage': cost_usage_pct
-                })
-                self._limits_cache[cache_key] = {
-                    'result': result,
-                    'expires_at': now + timedelta(seconds=30)
-                }
-                return result
-            except Exception as e:
-                logger.error(f"Error calculating usage percentages: {e}")
-                # Return basic success
-                return True, "Within limits", {}
         
-        except Exception as e:
-            logger.error(f"Unexpected error in check_usage_limits for {user_id}: {e}")
-            # STRICT: Fail closed - deny requests if subscription system fails
-            return False, f"Subscription check error: {str(e)}", {}
+        Returns:
+            (can_proceed, error_message, usage_info)
+        """
+        from .limit_validation import LimitValidator
+        validator = LimitValidator(self)
+        return validator.check_usage_limits(user_id, provider, tokens_requested, actual_provider_name)
     
     def estimate_tokens(self, text: str, provider: APIProvider) -> int:
         """Estimate token count for text based on provider."""
@@ -827,6 +586,16 @@ class PricingService:
         if not pricing:
             return None
         
+        # Return pricing info as dict
+        return {
+            'provider': pricing.provider.value,
+            'model_name': pricing.model_name,
+            'cost_per_input_token': pricing.cost_per_input_token,
+            'cost_per_output_token': pricing.cost_per_output_token,
+            'cost_per_request': pricing.cost_per_request,
+            'description': pricing.description
+        }
+    
     def check_comprehensive_limits(
         self, 
         user_id: str, 
@@ -835,6 +604,7 @@ class PricingService:
         """
         Comprehensive pre-flight validation that checks ALL limits before making ANY API calls.
         
+        Delegates to LimitValidator for actual validation logic.
         This prevents wasteful API calls by validating that ALL subsequent operations will succeed
         before making the first external API call.
         
@@ -850,202 +620,9 @@ class PricingService:
             (can_proceed, error_message, error_details)
             If can_proceed is False, error_message explains which limit would be exceeded
         """
-        try:
-            logger.info(f"[Pre-flight Check] 🔍 Starting comprehensive validation for user {user_id}")
-            logger.info(f"[Pre-flight Check] 📋 Validating {len(operations)} operation(s) before making any API calls")
-            
-            # Get current usage and limits once
-            current_period = self.get_current_billing_period(user_id) or datetime.now().strftime("%Y-%m")
-            usage = self.db.query(UsageSummary).filter(
-                UsageSummary.user_id == user_id,
-                UsageSummary.billing_period == current_period
-            ).first()
-            
-            if not usage:
-                # First usage this period, create summary
-                try:
-                    usage = UsageSummary(
-                        user_id=user_id,
-                        billing_period=current_period
-                    )
-                    self.db.add(usage)
-                    self.db.commit()
-                except Exception as create_error:
-                    logger.error(f"Error creating usage summary: {create_error}")
-                    self.db.rollback()
-                    return False, f"Failed to create usage summary: {str(create_error)}", {}
-            
-            # Get user limits
-            limits_dict = self.get_user_limits(user_id)
-            if not limits_dict:
-                # No subscription found - check for free tier
-                free_plan = self.db.query(SubscriptionPlan).filter(
-                    SubscriptionPlan.tier == SubscriptionTier.FREE,
-                    SubscriptionPlan.is_active == True
-                ).first()
-                if free_plan:
-                    limits_dict = self._plan_to_limits_dict(free_plan)
-                else:
-                    return False, "No subscription plan found. Please subscribe to a plan.", {}
-            
-            limits = limits_dict.get('limits', {})
-            
-            # Track cumulative usage across all operations
-            total_llm_calls = (
-                (usage.gemini_calls or 0) +
-                (usage.openai_calls or 0) +
-                (usage.anthropic_calls or 0) +
-                (usage.mistral_calls or 0)
-            )
-            total_llm_tokens = {}
-            total_images = usage.stability_calls or 0
-            
-            # Log current usage summary
-            logger.info(f"[Pre-flight Check] 📊 Current Usage Summary:")
-            logger.info(f"   └─ Total LLM Calls: {total_llm_calls}")
-            logger.info(f"   └─ Gemini Tokens: {usage.gemini_tokens or 0}, Mistral/HF Tokens: {usage.mistral_tokens or 0}")
-            logger.info(f"   └─ Image Calls: {total_images}")
-            
-            # Validate each operation
-            for op_idx, operation in enumerate(operations):
-                provider = operation.get('provider')
-                provider_name = provider.value if hasattr(provider, 'value') else str(provider)
-                tokens_requested = operation.get('tokens_requested', 0)
-                actual_provider_name = operation.get('actual_provider_name')
-                operation_type = operation.get('operation_type', 'unknown')
-                
-                display_provider_name = actual_provider_name or provider_name
-                
-                logger.info(f"[Pre-flight Check] ✅ Operation {op_idx + 1}/{len(operations)}: {operation_type}")
-                logger.info(f"   ├─ Provider: {display_provider_name} (enum: {provider_name})")
-                logger.info(f"   └─ Estimated Tokens: {tokens_requested}")
-                
-                # Check if this is an LLM provider
-                llm_providers = ['gemini', 'openai', 'anthropic', 'mistral']
-                is_llm_provider = provider_name in llm_providers
-                
-                # Check unified AI text generation limit for LLM providers
-                if is_llm_provider:
-                    ai_text_gen_limit = limits.get('ai_text_generation_calls', 0) or 0
-                    if ai_text_gen_limit == 0:
-                        # Fallback to provider-specific limit
-                        ai_text_gen_limit = limits.get(f"{provider_name}_calls", 0) or 0
-                    
-                    # Count this operation as an LLM call
-                    projected_total_llm_calls = total_llm_calls + 1
-                    
-                    if ai_text_gen_limit > 0 and projected_total_llm_calls > ai_text_gen_limit:
-                        error_info = {
-                            'current_calls': total_llm_calls,
-                            'limit': ai_text_gen_limit,
-                            'provider': display_provider_name,
-                            'operation_type': operation_type,
-                            'operation_index': op_idx
-                        }
-                        return False, f"AI text generation call limit would be exceeded. Would use {projected_total_llm_calls} of {ai_text_gen_limit} total AI text generation calls.", {
-                            'error_type': 'call_limit',
-                            'usage_info': error_info
-                        }
-                    
-                    # Check token limits for this provider
-                    # Use cumulative projected tokens from previous operations, or current from DB if first operation
-                    provider_tokens_key = f"{provider_name}_tokens"
-                    if provider_tokens_key in total_llm_tokens:
-                        # Use cumulative projected tokens from previous operations
-                        current_provider_tokens = total_llm_tokens[provider_tokens_key]
-                        logger.info(f"   └─ Using cumulative projected tokens: {current_provider_tokens}")
-                    else:
-                        # First operation for this provider - get current from database
-                        current_provider_tokens = getattr(usage, provider_tokens_key, 0) or 0
-                        total_llm_tokens[provider_tokens_key] = current_provider_tokens
-                        logger.info(f"   └─ Current tokens from DB: {current_provider_tokens}")
-                    
-                    token_limit = limits.get(provider_tokens_key, 0) or 0
-                    
-                    if token_limit > 0 and tokens_requested > 0:
-                        projected_tokens = current_provider_tokens + tokens_requested
-                        logger.info(f"   └─ Token Check: {current_provider_tokens} (current) + {tokens_requested} (requested) = {projected_tokens} (total) / {token_limit} (limit)")
-                        
-                        if projected_tokens > token_limit:
-                            usage_percentage = (projected_tokens / token_limit) * 100 if token_limit > 0 else 0
-                            error_info = {
-                                'current_tokens': current_provider_tokens,
-                                'requested_tokens': tokens_requested,
-                                'limit': token_limit,
-                                'provider': display_provider_name,
-                                'operation_type': operation_type,
-                                'operation_index': op_idx
-                            }
-                            error_msg = (
-                                f"Token limit exceeded for {display_provider_name} "
-                                f"({operation_type}). "
-                                f"Current: {current_provider_tokens}/{token_limit}, "
-                                f"Requested: {tokens_requested}, "
-                                f"Would exceed by: {projected_tokens - token_limit} tokens "
-                                f"({usage_percentage:.1f}% of limit)"
-                            )
-                            logger.error(f"[Pre-flight Check] ❌ BLOCKED: {error_msg}")
-                            return False, error_msg, {
-                                'error_type': 'token_limit',
-                                'usage_info': error_info
-                            }
-                        else:
-                            logger.info(f"   └─ ✅ Token limit check passed: {projected_tokens} <= {token_limit}")
-                    
-                    # Update cumulative counts for next operation
-                    total_llm_calls = projected_total_llm_calls
-                    total_llm_tokens[provider_tokens_key] += tokens_requested
-                    logger.info(f"   └─ Updated cumulative tokens for {display_provider_name}: {total_llm_tokens[provider_tokens_key]}")
-                
-                # Check image generation limits
-                elif provider == APIProvider.STABILITY:
-                    image_limit = limits.get('stability_calls', 0) or 0
-                    projected_images = total_images + 1
-                    
-                    if image_limit > 0 and projected_images > image_limit:
-                        error_info = {
-                            'current_images': total_images,
-                            'limit': image_limit,
-                            'provider': 'stability',
-                            'operation_type': operation_type,
-                            'operation_index': op_idx
-                        }
-                        return False, f"Image generation limit would be exceeded. Would use {projected_images} of {image_limit} images this billing period.", {
-                            'error_type': 'image_limit',
-                            'usage_info': error_info
-                        }
-                    
-                    total_images = projected_images
-                
-                # Check other provider-specific limits
-                else:
-                    provider_calls_key = f"{provider_name}_calls"
-                    current_provider_calls = getattr(usage, provider_calls_key, 0) or 0
-                    call_limit = limits.get(provider_calls_key, 0) or 0
-                    
-                    if call_limit > 0:
-                        projected_calls = current_provider_calls + 1
-                        if projected_calls > call_limit:
-                            error_info = {
-                                'current_calls': current_provider_calls,
-                                'limit': call_limit,
-                                'provider': display_provider_name,
-                                'operation_type': operation_type,
-                                'operation_index': op_idx
-                            }
-                            return False, f"API call limit would be exceeded for {display_provider_name}. Would use {projected_calls} of {call_limit} calls this billing period.", {
-                                'error_type': 'call_limit',
-                                'usage_info': error_info
-                            }
-            
-            # All checks passed
-            logger.info(f"[Pre-flight Check] ✅ All {len(operations)} operation(s) validated successfully")
-            logger.info(f"[Pre-flight Check] ✅ User {user_id} is cleared to proceed with API calls")
-            return True, None, None
-            
-        except Exception as e:
-            logger.error(f"[Pre-flight Check] Error during comprehensive limit check: {e}", exc_info=True)
-            return False, f"Failed to validate limits: {str(e)}", {}
+        from .limit_validation import LimitValidator
+        validator = LimitValidator(self)
+        return validator.check_comprehensive_limits(user_id, operations)
     
     def get_pricing_for_provider_model(self, provider: APIProvider, model_name: str) -> Optional[Dict[str, Any]]:
         """Get pricing configuration for a specific provider and model."""
diff --git a/backend/services/subscription/schema_utils.py b/backend/services/subscription/schema_utils.py
new file mode 100644
index 00000000..ac5bdcbc
--- /dev/null
+++ b/backend/services/subscription/schema_utils.py
@@ -0,0 +1,39 @@
+from typing import Set
+from sqlalchemy.orm import Session
+
+
+_checked_subscription_plan_columns: bool = False
+
+
+def ensure_subscription_plan_columns(db: Session) -> None:
+    """Ensure required columns exist on subscription_plans for runtime safety.
+
+    This is a defensive guard for environments where migrations have not yet
+    been applied. If columns are missing (e.g., exa_calls_limit), we add them
+    with a safe default so ORM queries do not fail.
+    """
+    global _checked_subscription_plan_columns
+    if _checked_subscription_plan_columns:
+        return
+
+    try:
+        # Discover existing columns
+        result = db.execute("PRAGMA table_info(subscription_plans)")
+        cols: Set[str] = {row[1] for row in result}
+
+        # Columns we may reference in models but might be missing in older DBs
+        required_columns = {
+            "exa_calls_limit": "INTEGER DEFAULT 0",
+        }
+
+        for col_name, ddl in required_columns.items():
+            if col_name not in cols:
+                db.execute(f"ALTER TABLE subscription_plans ADD COLUMN {col_name} {ddl}")
+        db.commit()
+    except Exception:
+        # Do not block app if pragma/alter fails; let normal errors surface
+        db.rollback()
+    finally:
+        _checked_subscription_plan_columns = True
+
+
diff --git a/backend/services/wix_service.py b/backend/services/wix_service.py
index 237d1238..29761455 100644
--- a/backend/services/wix_service.py
+++ b/backend/services/wix_service.py
@@ -7,6 +7,7 @@ Handles authentication, permission checking, and blog publishing to Wix websites
 import os
 import json
 import requests
+import uuid
 from typing import Dict, Any, Optional, List
 from loguru import logger
 from datetime import datetime, timedelta
@@ -19,6 +20,9 @@ from services.integrations.wix.media import WixMediaService
 from services.integrations.wix.utils import extract_meta_from_token, normalize_token_string, extract_member_id_from_access_token as utils_extract_member
 from services.integrations.wix.content import convert_content_to_ricos as ricos_builder
 from services.integrations.wix.auth import WixAuthService
+from services.integrations.wix.seo import build_seo_data
+from services.integrations.wix.ricos_converter import markdown_to_html, convert_via_wix_api
+from services.integrations.wix.blog_publisher import create_blog_post as publish_blog_post
 
 class WixService:
     """Service for interacting with Wix APIs"""
@@ -237,13 +241,35 @@ class WixService:
             logger.error(f"Failed to import image to Wix: {e}")
             raise
     
-    def convert_content_to_ricos(self, content: str, images: List[str] = None) -> Dict[str, Any]:
+    def convert_content_to_ricos(self, content: str, images: List[str] = None, 
+                                 use_wix_api: bool = False, access_token: str = None) -> Dict[str, Any]:
+        """
+        Convert markdown content to Ricos JSON format.
+        
+        Args:
+            content: Markdown content to convert
+            images: Optional list of image URLs
+            use_wix_api: If True, use Wix's official Ricos Documents API (requires access_token)
+            access_token: Wix access token (required if use_wix_api=True)
+            
+        Returns:
+            Ricos JSON document
+        """
+        if use_wix_api and access_token:
+            try:
+                return convert_via_wix_api(content, access_token, self.base_url)
+            except Exception as e:
+                logger.warning(f"Failed to convert via Wix API, falling back to custom parser: {e}")
+                # Fall back to custom parser
+        
+        # Use custom parser (current implementation)
         return ricos_builder(content, images)
     
+    
     def create_blog_post(self, access_token: str, title: str, content: str, 
                         cover_image_url: str = None, category_ids: List[str] = None,
                         tag_ids: List[str] = None, publish: bool = True, 
-                        member_id: str = None) -> Dict[str, Any]:
+                        member_id: str = None, seo_metadata: Dict[str, Any] = None) -> Dict[str, Any]:
         """
         Create and optionally publish a blog post on Wix
         
@@ -256,101 +282,33 @@ class WixService:
             tag_ids: Optional list of tag IDs
             publish: Whether to publish immediately or save as draft
             member_id: Required for third-party apps - the member ID of the post author
+            seo_metadata: Optional SEO metadata dict with fields like:
+                - seo_title: SEO optimized title
+                - meta_description: Meta description
+                - focus_keyword: Main keyword
+                - blog_tags: List of tag strings (for keywords)
+                - open_graph: Open Graph data
+                - canonical_url: Canonical URL
             
         Returns:
             Created blog post information
         """
-        if not member_id:
-            raise ValueError("memberId is required for third-party apps creating blog posts")
-        
-        headers = {
-            'Authorization': f'Bearer {access_token}',
-            'Content-Type': 'application/json'
-        }
-        
-        # Build valid Ricos rich content (minimum: one paragraph with text)
-        ricos_content = self.convert_content_to_ricos(content or "This is a post from ALwrity.", None)
-
-        # Minimal payload per Wix docs: title, memberId, and richContent
-        blog_data = {
-            'draftPost': {
-                'title': title,
-                'memberId': member_id,  # Required for third-party apps
-                'richContent': ricos_content,
-                'excerpt': (content or '').strip()[:200]
-            },
-            'publish': publish,
-            'fieldsets': ['URL']  # Simplified fieldsets
-        }
-        
-        # Add cover image if provided
-        if cover_image_url:
-            try:
-                media_id = self.import_image_to_wix(access_token, cover_image_url, f'Cover: {title}')
-                blog_data['draftPost']['media'] = {
-                    'wixMedia': {
-                        'image': {'id': media_id}
-                    },
-                    'displayed': True,
-                    'custom': True
-                }
-            except Exception as e:
-                logger.warning(f"Failed to import cover image: {e}")
-        
-        # Add categories if provided
-        if category_ids:
-            blog_data['draftPost']['categoryIds'] = category_ids
-        
-        # Add tags if provided
-        if tag_ids:
-            blog_data['draftPost']['tagIds'] = tag_ids
-        
-        try:
-            # Check what permissions we have in the token
-            logger.info("DEBUG: Checking token permissions...")
-            try:
-                import jwt
-                # Extract token string manually since _normalize_access_token doesn't exist
-                token_str = str(access_token)
-                if token_str and token_str.startswith('OauthNG.JWS.'):
-                    jwt_part = token_str[12:]
-                    payload = jwt.decode(jwt_part, options={"verify_signature": False, "verify_aud": False})
-                    logger.info(f"DEBUG: Full token payload: {payload}")
-                    
-                    # Check for permissions in various possible locations
-                    data_payload = payload.get('data', {})
-                    if isinstance(data_payload, str):
-                        try:
-                            data_payload = json.loads(data_payload)
-                        except:
-                            pass
-                    
-                    instance_data = data_payload.get('instance', {})
-                    permissions = instance_data.get('permissions', '')
-                    scopes = instance_data.get('scopes', [])
-                    meta_site_id = instance_data.get('metaSiteId')
-                    if isinstance(meta_site_id, str) and meta_site_id:
-                        headers['wix-site-id'] = meta_site_id
-                        logger.info(f"DEBUG: Added wix-site-id header: {meta_site_id}")
-                    logger.info(f"DEBUG: Token permissions: {permissions}")
-                    logger.info(f"DEBUG: Token scopes: {scopes}")
-                else:
-                    logger.info("DEBUG: Could not decode token for permission check")
-            except Exception as perm_e:
-                logger.warning(f"DEBUG: Failed to check permissions: {perm_e}")
-            
-            logger.info(f"DEBUG: Sending simplified blog data: {json.dumps(blog_data, indent=2)}")
-            extra_headers = {}
-            if 'wix-site-id' in headers:
-                extra_headers['wix-site-id'] = headers['wix-site-id']
-            result = self.blog_service.create_draft_post(access_token, blog_data, extra_headers or None)
-            logger.info(f"DEBUG: Create draft result: {result}")
-            return result
-        except requests.RequestException as e:
-            logger.error(f"Failed to create blog post: {e}")
-            if hasattr(e, 'response') and e.response is not None:
-                logger.error(f"Response body: {e.response.text}")
-            raise
+        return publish_blog_post(
+            blog_service=self.blog_service,
+            access_token=access_token,
+            title=title,
+            content=content,
+            member_id=member_id,
+            cover_image_url=cover_image_url,
+            category_ids=category_ids,
+            tag_ids=tag_ids,
+            publish=publish,
+            seo_metadata=seo_metadata,
+            import_image_func=self.import_image_to_wix,
+            lookup_categories_func=self.lookup_or_create_categories,
+            lookup_tags_func=self.lookup_or_create_tags,
+            base_url=self.base_url
+        )
     
     def get_blog_categories(self, access_token: str) -> List[Dict[str, Any]]:
         """
@@ -383,6 +341,138 @@ class WixService:
         except requests.RequestException as e:
             logger.error(f"Failed to get blog tags: {e}")
             raise
+    
+    def lookup_or_create_categories(self, access_token: str, category_names: List[str], 
+                                    extra_headers: Optional[Dict[str, str]] = None) -> List[str]:
+        """
+        Lookup existing categories by name or create new ones, return their IDs.
+        
+        Args:
+            access_token: Valid access token
+            category_names: List of category name strings
+            extra_headers: Optional extra headers (e.g., wix-site-id)
+            
+        Returns:
+            List of category UUIDs
+        """
+        if not category_names:
+            return []
+        
+        try:
+            # Get existing categories
+            existing_categories = self.blog_service.list_categories(access_token, extra_headers)
+            
+            # Create name -> ID mapping (case-insensitive)
+            category_map = {}
+            for cat in existing_categories:
+                cat_label = cat.get('label', '').strip()
+                cat_id = cat.get('id')
+                if cat_label and cat_id:
+                    category_map[cat_label.lower()] = cat_id
+            
+            category_ids = []
+            for category_name in category_names:
+                category_name_clean = str(category_name).strip()
+                if not category_name_clean:
+                    continue
+                
+                # Lookup existing category (case-insensitive)
+                category_id = category_map.get(category_name_clean.lower())
+                
+                if not category_id:
+                    # Create new category
+                    try:
+                        logger.info(f"Creating new category: {category_name_clean}")
+                        result = self.blog_service.create_category(
+                            access_token, 
+                            label=category_name_clean,
+                            extra_headers=extra_headers
+                        )
+                        new_category = result.get('category', {})
+                        category_id = new_category.get('id')
+                        if category_id:
+                            category_ids.append(category_id)
+                            # Update map to avoid duplicate creates
+                            category_map[category_name_clean.lower()] = category_id
+                            logger.info(f"Created category '{category_name_clean}' with ID: {category_id}")
+                    except Exception as create_error:
+                        logger.warning(f"Failed to create category '{category_name_clean}': {create_error}")
+                        # Continue with other categories
+                else:
+                    category_ids.append(category_id)
+                    logger.info(f"Found existing category '{category_name_clean}' with ID: {category_id}")
+            
+            return category_ids
+            
+        except requests.RequestException as e:
+            logger.error(f"Failed to lookup/create categories: {e}")
+            return []
+    
+    def lookup_or_create_tags(self, access_token: str, tag_names: List[str],
+                             extra_headers: Optional[Dict[str, str]] = None) -> List[str]:
+        """
+        Lookup existing tags by name or create new ones, return their IDs.
+        
+        Args:
+            access_token: Valid access token
+            tag_names: List of tag name strings
+            extra_headers: Optional extra headers (e.g., wix-site-id)
+            
+        Returns:
+            List of tag UUIDs
+        """
+        if not tag_names:
+            return []
+        
+        try:
+            # Get existing tags
+            existing_tags = self.blog_service.list_tags(access_token, extra_headers)
+            
+            # Create name -> ID mapping (case-insensitive)
+            tag_map = {}
+            for tag in existing_tags:
+                tag_label = tag.get('label', '').strip()
+                tag_id = tag.get('id')
+                if tag_label and tag_id:
+                    tag_map[tag_label.lower()] = tag_id
+            
+            tag_ids = []
+            for tag_name in tag_names:
+                tag_name_clean = str(tag_name).strip()
+                if not tag_name_clean:
+                    continue
+                
+                # Lookup existing tag (case-insensitive)
+                tag_id = tag_map.get(tag_name_clean.lower())
+                
+                if not tag_id:
+                    # Create new tag
+                    try:
+                        logger.info(f"Creating new tag: {tag_name_clean}")
+                        result = self.blog_service.create_tag(
+                            access_token,
+                            label=tag_name_clean,
+                            extra_headers=extra_headers
+                        )
+                        new_tag = result.get('tag', {})
+                        tag_id = new_tag.get('id')
+                        if tag_id:
+                            tag_ids.append(tag_id)
+                            # Update map to avoid duplicate creates
+                            tag_map[tag_name_clean.lower()] = tag_id
+                            logger.info(f"Created tag '{tag_name_clean}' with ID: {tag_id}")
+                    except Exception as create_error:
+                        logger.warning(f"Failed to create tag '{tag_name_clean}': {create_error}")
+                        # Continue with other tags
+                else:
+                    tag_ids.append(tag_id)
+                    logger.info(f"Found existing tag '{tag_name_clean}' with ID: {tag_id}")
+            
+            return tag_ids
+            
+        except requests.RequestException as e:
+            logger.error(f"Failed to lookup/create tags: {e}")
+            return []
 
     def publish_draft_post(self, access_token: str, draft_post_id: str) -> Dict[str, Any]:
         """
diff --git a/docs/RESEARCH_COMPONENT_INTEGRATION.md b/docs/RESEARCH_COMPONENT_INTEGRATION.md
new file mode 100644
index 00000000..8467fd2d
--- /dev/null
+++ b/docs/RESEARCH_COMPONENT_INTEGRATION.md
@@ -0,0 +1,335 @@
+# Research Component Integration Guide
+
+## Overview
+
+The modular Research component has been implemented as a standalone, testable wizard that can be integrated into the blog writer or used independently. This document outlines the architecture, usage, and integration steps.
+
+## Architecture
+
+### Backend Strategy Pattern
+
+The research service now supports multiple research modes through a strategy pattern:
+
+```python
+# Research modes
+- Basic: Quick keyword-focused analysis
+- Comprehensive: Full analysis with all components
+- Targeted: Customizable components based on config
+
+# Strategy implementation
+backend/services/blog_writer/research/research_strategies.py
+- ResearchStrategy (base class)
+- BasicResearchStrategy
+- ComprehensiveResearchStrategy
+- TargetedResearchStrategy
+```
+
+### Frontend Component Structure
+
+```
+frontend/src/components/Research/
+├── index.tsx                    # Main exports
+├── ResearchWizard.tsx          # Main wizard container
+├── steps/
+│   ├── StepKeyword.tsx         # Step 1: Keyword input
+│   ├── StepOptions.tsx         # Step 2: Mode selection
+│   ├── StepProgress.tsx        # Step 3: Progress display
+│   └── StepResults.tsx         # Step 4: Results display
+├── hooks/
+│   ├── useResearchWizard.ts   # Wizard state management
+│   └── useResearchExecution.ts # API calls and polling
+├── types/
+│   └── research.types.ts       # TypeScript interfaces
+└── utils/
+    └── researchUtils.ts        # Utility functions
+```
+
+## Test Page
+
+A dedicated test page is available at `/research-test` for testing the research wizard independently.
+
+**Features:**
+- Quick preset keywords for testing
+- Debug panel with JSON export
+- Performance metrics display
+- Cache state visualization
+
+## Usage
+
+### Standalone Usage
+
+```typescript
+import { ResearchWizard } from '../components/Research';
+
+<ResearchWizard
+  onComplete={(results) => {
+    console.log('Research complete:', results);
+  }}
+  onCancel={() => {
+    console.log('Cancelled');
+  }}
+  initialKeywords={['AI', 'marketing']}
+  initialIndustry="Technology"
+/>
+```
+
+### Integration with Blog Writer
+
+The component is designed to be easily integrated into the BlogWriter research phase:
+
+**Current Implementation:**
+- Uses CopilotKit sidebar for research input
+- Displays results in `ResearchResults` component
+- Manual fallback via `ManualResearchForm`
+
+**Proposed Integration:**
+Replace the CopilotKit/manual form with the wizard:
+
+```typescript
+// In BlogWriter.tsx
+{currentPhase === 'research' && (
+  <ResearchWizard
+    onComplete={(results) => setResearch(results)}
+    onCancel={() => navigate('blog-writer')}
+  />
+)}
+```
+
+## Backend API Changes
+
+### New Models
+
+The `BlogResearchRequest` model now supports:
+
+```python
+class BlogResearchRequest(BaseModel):
+    keywords: List[str]
+    topic: Optional[str] = None
+    industry: Optional[str] = None
+    target_audience: Optional[str] = None
+    tone: Optional[str] = None
+    word_count_target: Optional[int] = 1500
+    persona: Optional[PersonaInfo] = None
+    research_mode: Optional[ResearchMode] = ResearchMode.BASIC  # NEW
+    config: Optional[ResearchConfig] = None  # NEW
+```
+
+### Backward Compatibility
+
+The API remains backward compatible:
+- If `research_mode` is not provided, defaults to `BASIC`
+- If `config` is not provided, defaults to standard configuration
+- Existing requests continue to work unchanged
+
+## Research Modes
+
+### Basic Mode
+- Quick keyword analysis
+- Primary & secondary keywords
+- Current trends overview
+- Top 5 content angles
+- Key statistics
+
+### Comprehensive Mode
+- All basic features plus:
+- Expert quotes & opinions
+- Competitor analysis
+- Market forecasts
+- Best practices & case studies
+- Content gaps identification
+
+### Targeted Mode
+- Selectable components:
+  - Statistics
+  - Expert quotes
+  - Competitors
+  - Trends
+  - Always includes: Keywords & content angles
+
+## Configuration Options
+
+### ResearchConfig Model
+
+```python
+class ResearchConfig(BaseModel):
+    mode: ResearchMode = ResearchMode.BASIC
+    date_range: Optional[DateRange] = None
+    source_types: List[SourceType] = []
+    max_sources: int = 10
+    include_statistics: bool = True
+    include_expert_quotes: bool = True
+    include_competitors: bool = True
+    include_trends: bool = True
+```
+
+### Date Range Options
+- `last_week`
+- `last_month`
+- `last_3_months`
+- `last_6_months`
+- `last_year`
+- `all_time`
+
+### Source Types
+- `web` - Web articles
+- `academic` - Academic papers
+- `news` - News articles
+- `industry` - Industry reports
+- `expert` - Expert opinions
+
+## Caching
+
+The research component uses the existing cache infrastructure:
+- Cache keys include research mode
+- Cache is shared across basic/comprehensive/targeted modes
+- Cache invalidation handled automatically
+
+## Testing
+
+### Test the Wizard
+
+1. Navigate to `/research-test`
+2. Use quick presets or enter custom keywords
+3. Select research mode
+4. Monitor progress
+5. Review results
+6. Export JSON for analysis
+
+### Integration Testing
+
+To test integration with BlogWriter:
+
+1. Start backend: `python start_alwrity_backend.py`
+2. Navigate to `/blog-writer` (current implementation)
+3. Or navigate to `/research-test` (new wizard)
+4. Compare results and UI
+
+## Migration Path
+
+### Phase 1: Parallel Testing (Current)
+- `/research-test` - New wizard available
+- `/blog-writer` - Current implementation unchanged
+- Users can test both
+
+### Phase 2: Integration
+1. Add wizard as option in BlogWriter
+2. A/B test user preference
+3. Monitor performance metrics
+
+### Phase 3: Replacement (Optional)
+1. Replace CopilotKit/manual form with wizard
+2. Remove old implementation
+3. Update documentation
+
+## API Endpoints
+
+All existing endpoints remain unchanged:
+
+```
+POST /api/blog/research/start
+- Supports new research_mode and config parameters
+- Backward compatible with existing requests
+
+GET /api/blog/research/status/{task_id}
+- No changes required
+```
+
+## Benefits
+
+1. **Modularity**: Component works standalone
+2. **Testability**: Dedicated test page for experimentation
+3. **Backward Compatibility**: Existing functionality unchanged
+4. **Progressive Enhancement**: Can add features incrementally
+5. **Reusability**: Can be used in other parts of the app
+
+## Future Enhancements
+
+Potential future improvements:
+
+1. **Multi-stage Research**: Sequential research with refinement
+2. **Source Quality Validation**: Advanced credibility scoring
+3. **Interactive Query Builder**: Dynamic search refinement
+4. **Advanced Prompting**: Few-shot examples, reasoning chains
+5. **Custom Strategy Plugins**: User-defined research strategies
+
+## Troubleshooting
+
+### Research Results Not Showing
+
+Check:
+1. Backend logs for API errors
+2. Network tab for failed requests
+3. Browser console for JavaScript errors
+4. Verify user authentication
+
+### Cache Issues
+
+Clear cache:
+```typescript
+import { researchCache } from '../services/researchCache';
+researchCache.clearCache();
+```
+
+### Type Errors
+
+Ensure all imports are correct:
+```typescript
+import { 
+  ResearchWizard,
+  useResearchWizard,
+  WizardState 
+} from '../components/Research';
+
+import {
+  BlogResearchRequest,
+  BlogResearchResponse,
+  ResearchMode,
+  ResearchConfig
+} from '../services/blogWriterApi';
+```
+
+## Examples
+
+### Basic Integration
+
+```typescript
+import { ResearchWizard } from './components/Research';
+import { BlogResearchResponse } from './services/blogWriterApi';
+
+const MyComponent: React.FC = () => {
+  const [results, setResults] = useState<BlogResearchResponse | null>(null);
+
+  return (
+    <ResearchWizard
+      onComplete={(res) => setResults(res)}
+      onCancel={() => console.log('Cancelled')}
+    />
+  );
+};
+```
+
+### Advanced Integration with Custom Config
+
+```typescript
+const request: BlogResearchRequest = {
+  keywords: ['AI', 'automation'],
+  industry: 'Technology',
+  research_mode: 'targeted',
+  config: {
+    mode: 'targeted',
+    include_statistics: true,
+    include_competitors: true,
+    include_trends: false,
+    max_sources: 20,
+  }
+};
+```
+
+## Support
+
+For issues or questions:
+1. Check this documentation
+2. Review test page examples
+3. Inspect backend logs
+4. Check frontend console
+
diff --git a/docs/RESEARCH_WIZARD_IMPLEMENTATION.md b/docs/RESEARCH_WIZARD_IMPLEMENTATION.md
new file mode 100644
index 00000000..5b980d30
--- /dev/null
+++ b/docs/RESEARCH_WIZARD_IMPLEMENTATION.md
@@ -0,0 +1,346 @@
+# Research Wizard Implementation Summary
+
+## Implementation Complete
+
+A modular, pluggable research component has been successfully implemented with wizard-based UI that can be tested independently and integrated into the blog writer.
+
+---
+
+## Backend Implementation
+
+### 1. Research Models (blog_models.py)
+
+**New Enums:**
+- `ResearchMode`: `BASIC`, `COMPREHENSIVE`, `TARGETED`
+- `SourceType`: `WEB`, `ACADEMIC`, `NEWS`, `INDUSTRY`, `EXPERT`
+- `DateRange`: `LAST_WEEK` through `ALL_TIME`
+
+**New Models:**
+```python
+class ResearchConfig(BaseModel):
+    mode: ResearchMode = ResearchMode.BASIC
+    date_range: Optional[DateRange] = None
+    source_types: List[SourceType] = []
+    max_sources: int = 10
+    include_statistics: bool = True
+    include_expert_quotes: bool = True
+    include_competitors: bool = True
+    include_trends: bool = True
+```
+
+**Enhanced BlogResearchRequest:**
+- Added `research_mode: Optional[ResearchMode]`
+- Added `config: Optional[ResearchConfig]`
+- **Backward compatible** - defaults to existing behavior
+
+### 2. Strategy Pattern (research_strategies.py)
+
+**New file:** `backend/services/blog_writer/research/research_strategies.py`
+
+**Three Strategy Classes:**
+1. **BasicResearchStrategy**: Quick keyword-focused analysis
+2. **ComprehensiveResearchStrategy**: Full analysis with all components
+3. **TargetedResearchStrategy**: Customizable components based on config
+
+**Factory Function:**
+```python
+get_strategy_for_mode(mode: ResearchMode) -> ResearchStrategy
+```
+
+### 3. Service Integration (research_service.py)
+
+**Key Changes:**
+- Imports strategy factory and models
+- Uses strategy pattern in both `research()` and `research_with_progress()` methods
+- Automatically selects strategy based on `research_mode`
+- Backward compatible - defaults to BASIC if not specified
+
+**Line Changes:**
+```python
+# Lines 88-96: Determine research mode and get appropriate strategy
+research_mode = request.research_mode or ResearchMode.BASIC
+config = request.config or ResearchConfig(mode=research_mode)
+strategy = get_strategy_for_mode(research_mode)
+
+logger.info(f"Using research mode: {research_mode.value}")
+
+# Build research prompt based on strategy
+research_prompt = strategy.build_research_prompt(topic, industry, target_audience, config)
+```
+
+---
+
+## Frontend Implementation
+
+### 4. Component Structure
+
+**New Directory:** `frontend/src/components/Research/`
+
+```
+Research/
+├── index.tsx                          # Main exports
+├── ResearchWizard.tsx                # Main wizard container
+├── steps/
+│   ├── StepKeyword.tsx               # Step 1: Keyword input
+│   ├── StepOptions.tsx               # Step 2: Mode selection (3 cards)
+│   ├── StepProgress.tsx              # Step 3: Progress display
+│   └── StepResults.tsx               # Step 4: Results display
+├── hooks/
+│   ├── useResearchWizard.ts          # Wizard state management
+│   └── useResearchExecution.ts       # API calls and polling
+├── types/
+│   └── research.types.ts             # TypeScript interfaces
+├── utils/
+│   └── researchUtils.ts              # Utility functions
+└── integrations/
+    └── BlogWriterAdapter.tsx         # Blog writer integration adapter
+```
+
+### 5. Wizard Components
+
+**ResearchWizard.tsx:**
+- Main container with progress bar
+- Step indicators (Setup → Options → Research → Results)
+- Navigation footer with Back/Next buttons
+- Responsive layout
+
+**StepKeyword.tsx:**
+- Keywords textarea
+- Industry dropdown (16 options)
+- Target audience input
+- Validation for keyword requirements
+
+**StepOptions.tsx:**
+- Three mode cards (Basic, Comprehensive, Targeted)
+- Visual selection feedback
+- Feature lists per mode
+- Hover effects
+
+**StepProgress.tsx:**
+- Real-time progress updates
+- Progress messages display
+- Cancel button
+- Auto-advance to results on completion
+
+**StepResults.tsx:**
+- Displays research results using existing `ResearchResults` component
+- Export JSON button
+- Start new research button
+
+### 6. Hooks
+
+**useResearchWizard.ts:**
+- State management for wizard steps
+- localStorage persistence
+- Step navigation (next/back)
+- Validation per step
+- Reset functionality
+
+**useResearchExecution.ts:**
+- Research execution via API
+- Cache checking
+- Polling integration
+- Error handling
+- Progress tracking
+
+### 7. Test Page (ResearchTest.tsx)
+
+**Location:** `frontend/src/pages/ResearchTest.tsx`  
+**Route:** `/research-test`
+
+**Features:**
+- Quick preset buttons (3 samples)
+- Debug panel with JSON export
+- Performance metrics display
+- Cache state visualization
+- Research statistics summary
+
+**Sample Presets:**
+1. AI Marketing Tools
+2. Small Business SEO
+3. Content Strategy
+
+### 8. Type Definitions
+
+**research.types.ts:**
+- `WizardState`
+- `WizardStepProps`
+- `ResearchWizardProps`
+- `ModeCardInfo`
+
+**blogWriterApi.ts:**
+- `ResearchMode` type union
+- `SourceType` type union
+- `DateRange` type union
+- `ResearchConfig` interface
+- Updated `BlogResearchRequest` interface
+
+---
+
+## Integration
+
+### 9. Blog Writer API (blogWriterApi.ts)
+
+**Enhanced Interface:**
+```typescript
+export interface BlogResearchRequest {
+  keywords: string[];
+  topic?: string;
+  industry?: string;
+  target_audience?: string;
+  tone?: string;
+  word_count_target?: number;
+  persona?: PersonaInfo;
+  research_mode?: ResearchMode;      // NEW
+  config?: ResearchConfig;           // NEW
+}
+```
+
+### 10. App Routing (App.tsx)
+
+**New Route:**
+```typescript
+<Route path="/research-test" element={<ResearchTest />} />
+```
+
+### 11. Integration Adapter
+
+**BlogWriterAdapter.tsx:**
+- Wrapper component for easy integration
+- Usage examples included
+- Clean interface for BlogWriter
+
+---
+
+## Documentation
+
+### 12. Integration Guide
+
+**File:** `docs/RESEARCH_COMPONENT_INTEGRATION.md`
+
+**Contents:**
+- Architecture overview
+- Usage examples
+- Backend API details
+- Research modes explained
+- Configuration options
+- Testing instructions
+- Migration path
+- Troubleshooting guide
+
+---
+
+## Key Features
+
+### Research Modes
+
+**Basic Mode:**
+- Quick keyword analysis
+- Primary & secondary keywords
+- Trends overview
+- Top 5 content angles
+- Key statistics
+
+**Comprehensive Mode:**
+- All basic features
+- Expert quotes & opinions
+- Competitor analysis
+- Market forecasts
+- Best practices & case studies
+- Content gaps identification
+
+**Targeted Mode:**
+- Selectable components
+- Customizable filters
+- Date range options
+- Source type filtering
+
+### User Experience
+
+1. **Step-by-step wizard** with clear progress
+2. **Visual mode selection** with cards
+3. **Real-time progress** with live updates
+4. **Comprehensive results** with export capability
+5. **Error handling** with retry options
+6. **Cache integration** for instant results
+
+### Developer Experience
+
+1. **Modular architecture** - standalone components
+2. **Type safety** - full TypeScript interfaces
+3. **Reusable hooks** - state and execution management
+4. **Test page** - isolated testing environment
+5. **Documentation** - comprehensive guides
+
+---
+
+## Testing
+
+### Quick Test
+
+1. Navigate to `http://localhost:3000/research-test`
+2. Click "AI Marketing Tools" preset
+3. Select "Comprehensive" mode
+4. Watch progress updates
+5. Review results with export
+
+### Integration Test
+
+1. Compare `/research-test` wizard UI
+2. Compare `/blog-writer` current UI
+3. Test both research workflows
+4. Verify caching works across both
+
+---
+
+## Backward Compatibility
+
+- Existing API calls continue working
+- No breaking changes to BlogWriter
+- Optional parameters default to current behavior
+- Cache infrastructure shared
+- All existing features preserved
+
+---
+
+## File Summary
+
+**Backend (4 files):**
+- Modified: `blog_models.py`, `research_service.py`
+- Created: `research_strategies.py`
+
+**Frontend (13 files):**
+- Created: `ResearchWizard.tsx`, 4 step components, 2 hooks, types, utils, adapter, test page
+- Modified: `App.tsx`, `blogWriterApi.ts`
+
+**Documentation (2 files):**
+- Created: `RESEARCH_COMPONENT_INTEGRATION.md`, `RESEARCH_WIZARD_IMPLEMENTATION.md`
+
+---
+
+## Next Steps
+
+1. ✅ **Test the wizard** at `/research-test`
+2. ✅ **Review integration guide** in docs
+3. ⏳ **Integrate into BlogWriter** using adapter (optional)
+4. ⏳ **Gather user feedback** on wizard vs CopilotKit UI
+5. ⏳ **Add more presets** if needed
+
+---
+
+## Benefits Delivered
+
+- Modular & Pluggable: Standalone component
+- Testable: Dedicated test page
+- Backward Compatible: No breaking changes
+- Reusable: Can be used anywhere in the app
+- Extensible: Easy to add new modes or features
+- Documented: Comprehensive guides
+- Type Safe: Full TypeScript support
+- Production Ready: No linting errors
+
+---
+
+Implementation Date: Current Session  
+Status: Complete & Ready for Testing
+
diff --git a/docs/WIX_SEO_METADATA_COMPLETE.md b/docs/WIX_SEO_METADATA_COMPLETE.md
new file mode 100644
index 00000000..de6ee605
--- /dev/null
+++ b/docs/WIX_SEO_METADATA_COMPLETE.md
@@ -0,0 +1,150 @@
+# Complete Wix SEO Metadata Implementation
+
+## 📊 SEO Metadata Generated vs Posted
+
+### ✅ FULLY POSTED TO WIX
+
+#### 1. **SEO Keywords** (in `seoData.settings.keywords`)
+- ✅ `focus_keyword` → Main keyword (`isMain: true`)
+- ✅ `blog_tags` → Additional keywords (`isMain: false`)
+- ✅ `social_hashtags` → Additional keywords (`isMain: false`)
+
+#### 2. **Meta Tags** (in `seoData.tags`)
+- ✅ `meta_description` → `<meta name="description">`
+- ✅ `seo_title` → `<meta name="title">`
+
+#### 3. **Open Graph Tags** (in `seoData.tags`)
+- ✅ `open_graph.title` → `og:title`
+- ✅ `open_graph.description` → `og:description`
+- ✅ `open_graph.image` → `og:image` (HTTP/HTTPS URLs only)
+- ✅ `og:type` → Always set to `article`
+- ✅ `open_graph.url` or `canonical_url` → `og:url`
+
+#### 4. **Twitter Card Tags** (in `seoData.tags`)
+- ✅ `twitter_card.title` → `twitter:title`
+- ✅ `twitter_card.description` → `twitter:description`
+- ✅ `twitter_card.image` → `twitter:image` (HTTP/HTTPS URLs only)
+- ✅ `twitter_card.card` → `twitter:card` (default: `summary_large_image`)
+
+#### 5. **Canonical URL** (in `seoData.tags`)
+- ✅ `canonical_url` → `<link rel="canonical">`
+
+#### 6. **Blog Categories** (in `draftPost.categoryIds`)
+- ✅ `blog_categories` → Lookup/create categories → `categoryIds` (UUIDs)
+- **Implementation**: `lookup_or_create_categories()` method
+- **Behavior**: Case-insensitive lookup, auto-create if missing
+
+#### 7. **Blog Tags** (in `draftPost.tagIds`)
+- ✅ `blog_tags` → Lookup/create tags → `tagIds` (UUIDs)
+- **Implementation**: `lookup_or_create_tags()` method
+- **Behavior**: Case-insensitive lookup, auto-create if missing
+- **Note**: `blog_tags` are also used in SEO keywords, but separately as post tags
+
+### ❌ NOT POSTED (Optional/Future)
+
+1. **JSON-LD Structured Data** (`json_ld_schema`)
+   - **Reason**: Wix doesn't support JSON-LD in backend API
+   - **Solution**: Would require frontend implementation using `@wix/site-seo` package
+   - **Status**: Not implemented (would need to be added to Wix site code)
+
+2. **URL Slug** (`url_slug`)
+   - **Reason**: Wix auto-generates URLs from title
+   - **Status**: Could be implemented if Wix API supports custom slugs
+
+3. **Reading Time** (`reading_time`)
+   - **Reason**: Metadata only, not part of Wix blog post structure
+   - **Status**: Not applicable
+
+4. **Optimization Score** (`optimization_score`)
+   - **Reason**: Internal metadata for ALwrity, not Wix field
+   - **Status**: Not applicable
+
+## 🔄 Conversion Methods
+
+### Markdown to Ricos Conversion
+
+**Primary Method**: Wix Official Ricos Documents API
+- **Endpoint**: Tries multiple paths to find correct endpoint
+- **Benefits**: Official conversion, handles all edge cases
+- **Fallback**: Custom parser if API unavailable
+
+**Fallback Method**: Custom Markdown Parser
+- **Location**: `backend/services/integrations/wix/content.py`
+- **Supports**: Headings, paragraphs, lists, bold, italic, links, images, blockquotes
+
+## 📋 Complete Post Structure
+
+When publishing to Wix, the blog post includes:
+
+```json
+{
+  "draftPost": {
+    "title": "SEO optimized title",
+    "memberId": "author-member-id",
+    "richContent": { /* Ricos JSON document */ },
+    "excerpt": "First 200 chars of content",
+    "categoryIds": ["uuid1", "uuid2"],  // From blog_categories
+    "tagIds": ["uuid1", "uuid2"],        // From blog_tags
+    "media": { /* Cover image if provided */ },
+    "seoData": {
+      "settings": {
+        "keywords": [
+          { "term": "main keyword", "isMain": true },
+          { "term": "tag1", "isMain": false },
+          { "term": "tag2", "isMain": false }
+        ]
+      },
+      "tags": [
+        { "type": "meta", "props": { "name": "description", "content": "..." } },
+        { "type": "meta", "props": { "name": "title", "content": "..." } },
+        { "type": "meta", "props": { "property": "og:title", "content": "..." } },
+        { "type": "meta", "props": { "property": "og:description", "content": "..." } },
+        { "type": "meta", "props": { "property": "og:image", "content": "..." } },
+        { "type": "meta", "props": { "property": "og:type", "content": "article" } },
+        { "type": "meta", "props": { "property": "og:url", "content": "..." } },
+        { "type": "meta", "props": { "name": "twitter:title", "content": "..." } },
+        { "type": "meta", "props": { "name": "twitter:description", "content": "..." } },
+        { "type": "meta", "props": { "name": "twitter:image", "content": "..." } },
+        { "type": "meta", "props": { "name": "twitter:card", "content": "summary_large_image" } },
+        { "type": "link", "props": { "rel": "canonical", "href": "..." } }
+      ]
+    }
+  },
+  "publish": true
+}
+```
+
+## ✅ Implementation Status
+
+### Fully Implemented ✅
+- SEO keywords (main + additional)
+- Meta description and title
+- Open Graph tags (all standard fields)
+- Twitter Card tags (all standard fields)
+- Canonical URL
+- **Blog categories** (lookup/create)
+- **Blog tags** (lookup/create)
+- Wix Ricos API integration (with fallback)
+
+### Partially Implemented ⚠️
+- Image handling (only HTTP/HTTPS URLs, base64 skipped)
+
+### Not Implemented ❌
+- JSON-LD structured data (requires frontend)
+- URL slug customization
+- Reading time (not applicable)
+- Optimization score (not applicable)
+
+## 🎯 Summary
+
+**All major SEO metadata fields are now being posted to Wix:**
+- ✅ Keywords
+- ✅ Meta tags
+- ✅ Open Graph
+- ✅ Twitter Cards
+- ✅ Canonical URL
+- ✅ Categories (auto-lookup/create)
+- ✅ Tags (auto-lookup/create)
+
+The only missing piece is JSON-LD structured data, which requires frontend implementation in the Wix site code using the `@wix/site-seo` package.
+
diff --git a/docs/WIX_SEO_METADATA_REVIEW.md b/docs/WIX_SEO_METADATA_REVIEW.md
new file mode 100644
index 00000000..4d6f7a19
--- /dev/null
+++ b/docs/WIX_SEO_METADATA_REVIEW.md
@@ -0,0 +1,102 @@
+# Wix SEO Metadata Review
+
+## SEO Metadata We Generate (`BlogSEOMetadataResponse`)
+
+### Available Fields:
+1. ✅ **seo_title** - SEO optimized title
+2. ✅ **meta_description** - Meta description
+3. ✅ **url_slug** - URL slug for the blog post
+4. ✅ **blog_tags** - Array of tag strings (NOW being used for Wix post tags via lookup/create)
+5. ✅ **blog_categories** - Array of category strings (NOW being used for Wix post categories via lookup/create)
+6. ✅ **social_hashtags** - Hashtags for social media
+7. ✅ **open_graph** - Open Graph metadata object:
+   - title
+   - description
+   - image
+   - url
+   - type
+8. ✅ **twitter_card** - Twitter Card metadata object:
+   - title
+   - description
+   - image
+   - card (type)
+9. ✅ **canonical_url** - Canonical URL
+10. ✅ **focus_keyword** - Main SEO keyword
+11. ❌ **json_ld_schema** - JSON-LD structured data (NOT being posted - would need frontend implementation)
+12. ❌ **schema** - Legacy schema field (NOT being used)
+13. ❌ **reading_time** - Estimated reading time (NOT being posted)
+14. ❌ **optimization_score** - SEO optimization score (NOT being posted)
+15. ❌ **generated_at** - Generation timestamp (NOT being posted)
+
+## What We're Currently Posting to Wix
+
+### ✅ Posted via `seoData`:
+- **Keywords** (from `focus_keyword`, `blog_tags`, `social_hashtags`)
+  - Main keyword: `focus_keyword` → `isMain: true`
+  - Additional keywords: `blog_tags` and `social_hashtags` → `isMain: false`
+- **Meta Tags**:
+  - `meta description` → `<meta name="description">`
+  - `seo_title` → `<meta name="title">`
+- **Open Graph Tags**:
+  - `og:title`, `og:description`, `og:image`, `og:type`, `og:url`
+- **Twitter Card Tags**:
+  - `twitter:title`, `twitter:description`, `twitter:image`, `twitter:card`
+- **Canonical URL**:
+  - `<link rel="canonical">`
+
+### ✅ NOW Being Posted (Recently Implemented):
+
+1. **Blog Categories** (`blog_categories`)
+   - ✅ **Implemented**: `lookup_or_create_categories()` method
+   - ✅ **Behavior**: Case-insensitive lookup, auto-create if missing
+   - ✅ **Result**: Categories from SEO metadata are posted as `categoryIds` (UUIDs)
+
+2. **Blog Tags** (`blog_tags` for post organization)
+   - ✅ **Implemented**: `lookup_or_create_tags()` method
+   - ✅ **Behavior**: Case-insensitive lookup, auto-create if missing
+   - ✅ **Result**: Tags from SEO metadata are posted as `tagIds` (UUIDs)
+   - **Note**: `blog_tags` are used BOTH for SEO keywords AND for Wix post tags
+
+3. **JSON-LD Structured Data** (`json_ld_schema`)
+   - **Issue**: Wix doesn't support JSON-LD in backend API
+   - **Solution**: Would need frontend implementation using `@wix/site-seo` package
+   - **Status**: Not implemented
+
+4. **URL Slug** (`url_slug`)
+   - **Issue**: Not being passed to Wix
+   - **Status**: Wix generates URL automatically, but we could potentially set it
+
+## Implementation Status
+
+### ✅ Fully Implemented:
+- SEO keywords in `seoData.settings.keywords`
+- Meta description tag
+- SEO title tag
+- Open Graph tags (title, description, image, type, url)
+- Twitter Card tags (title, description, image, card type)
+- Canonical URL link tag
+
+### ✅ Fully Implemented:
+- **Blog Categories**: Auto-lookup/create from `blog_categories`
+- **Blog Tags**: Auto-lookup/create from `blog_tags`
+- **Wix Ricos API Integration**: Uses official Wix API with fallback to custom parser
+
+### ❌ Not Implemented (Optional):
+- JSON-LD structured data (frontend only - requires `@wix/site-seo` package)
+- URL slug setting (Wix auto-generates URLs)
+- Reading time (metadata only, not applicable)
+- Optimization score (metadata only, not applicable)
+
+## Summary
+
+✅ **All major SEO metadata is now being posted to Wix:**
+- SEO keywords (main + additional)
+- Meta tags (description, title)
+- Open Graph tags (title, description, image, type, url)
+- Twitter Card tags (title, description, image, card type)
+- Canonical URL
+- **Blog Categories** (auto-lookup/create)
+- **Blog Tags** (auto-lookup/create)
+
+The only missing piece is JSON-LD structured data, which requires frontend implementation in the Wix site code using `@wix/site-seo` package (not a backend concern).
+
diff --git a/frontend/src/App.tsx b/frontend/src/App.tsx
index 53746948..f2f0cc7a 100644
--- a/frontend/src/App.tsx
+++ b/frontend/src/App.tsx
@@ -17,13 +17,16 @@ import WixCallbackPage from './components/WixCallbackPage/WixCallbackPage';
 import WordPressCallbackPage from './components/WordPressCallbackPage/WordPressCallbackPage';
 import BingCallbackPage from './components/BingCallbackPage/BingCallbackPage';
 import BingAnalyticsStorage from './components/BingAnalyticsStorage/BingAnalyticsStorage';
+import ResearchTest from './pages/ResearchTest';
 import ProtectedRoute from './components/shared/ProtectedRoute';
 import GSCAuthCallback from './components/SEODashboard/components/GSCAuthCallback';
 import Landing from './components/Landing/Landing';
 import ErrorBoundary from './components/shared/ErrorBoundary';
 import ErrorBoundaryTest from './components/shared/ErrorBoundaryTest';
+import CopilotKitDegradedBanner from './components/shared/CopilotKitDegradedBanner';
 import { OnboardingProvider } from './contexts/OnboardingContext';
 import { SubscriptionProvider, useSubscription } from './contexts/SubscriptionContext';
+import { CopilotKitHealthProvider } from './contexts/CopilotKitHealthContext';
 
 import { setAuthTokenGetter } from './api/client';
 import { useOnboarding } from './contexts/OnboardingContext';
@@ -397,6 +400,7 @@ const App: React.FC = () => {
                 <Route path="/linkedin-writer" element={<ProtectedRoute><LinkedInWriter /></ProtectedRoute>} />
                 <Route path="/blog-writer" element={<ProtectedRoute><BlogWriter /></ProtectedRoute>} />
                 <Route path="/pricing" element={<PricingPage />} />
+                <Route path="/research-test" element={<ResearchTest />} />
                 <Route path="/wix-test" element={<WixTestPage />} />
                 <Route path="/wix-test-direct" element={<WixTestPage />} />
                 <Route path="/wix/callback" element={<WixCallbackPage />} />
@@ -411,14 +415,57 @@ const App: React.FC = () => {
 
     // Only wrap with CopilotKit if we have a valid key
     if (copilotApiKey && copilotApiKey.trim()) {
+      // Enhanced error handler that updates health context
+      const handleCopilotKitError = (e: any) => {
+        console.error("CopilotKit Error:", e);
+        
+        // Try to get health context if available
+        // We'll use a custom event to notify health context since we can't access it directly here
+        const errorMessage = e?.error?.message || e?.message || 'CopilotKit error occurred';
+        const errorType = errorMessage.toLowerCase();
+        
+        // Differentiate between fatal and transient errors
+        const isFatalError = 
+          errorType.includes('cors') ||
+          errorType.includes('ssl') ||
+          errorType.includes('certificate') ||
+          errorType.includes('403') ||
+          errorType.includes('forbidden') ||
+          errorType.includes('ERR_CERT_COMMON_NAME_INVALID');
+        
+        // Dispatch event for health context to listen to
+        window.dispatchEvent(new CustomEvent('copilotkit-error', {
+          detail: {
+            error: e,
+            errorMessage,
+            isFatal: isFatalError,
+          }
+        }));
+      };
+
       return (
-        <CopilotKit 
-          publicApiKey={copilotApiKey}
-          showDevConsole={false}
-          onError={(e) => console.error("CopilotKit Error:", e)}
+        <ErrorBoundary 
+          context="CopilotKit" 
+          showDetails={process.env.NODE_ENV === 'development'}
+          fallback={
+            <Box sx={{ p: 3, textAlign: 'center' }}>
+              <Typography variant="h6" color="warning" gutterBottom>
+                Chat Unavailable
+              </Typography>
+              <Typography variant="body2" color="textSecondary">
+                CopilotKit encountered an error. The app continues to work with manual controls.
+              </Typography>
+            </Box>
+          }
         >
-          {appContent}
-        </CopilotKit>
+          <CopilotKit 
+            publicApiKey={copilotApiKey}
+            showDevConsole={false}
+            onError={handleCopilotKitError}
+          >
+            {appContent}
+          </CopilotKit>
+        </ErrorBoundary>
       );
     }
 
@@ -426,6 +473,9 @@ const App: React.FC = () => {
     return appContent;
   };
 
+  // Determine initial health status based on whether CopilotKit key is available
+  const hasCopilotKitKey = copilotApiKey && copilotApiKey.trim();
+  
   return (
     <ErrorBoundary 
       context="Application Root"
@@ -439,7 +489,10 @@ const App: React.FC = () => {
       <ClerkProvider publishableKey={clerkPublishableKey}>
         <SubscriptionProvider>
           <OnboardingProvider>
-            {renderApp()}
+            <CopilotKitHealthProvider initialHealthStatus={!!hasCopilotKitKey}>
+              <CopilotKitDegradedBanner />
+              {renderApp()}
+            </CopilotKitHealthProvider>
           </OnboardingProvider>
         </SubscriptionProvider>
       </ClerkProvider>
diff --git a/frontend/src/api/client.ts b/frontend/src/api/client.ts
index 0bde5b6d..5935be8b 100644
--- a/frontend/src/api/client.ts
+++ b/frontend/src/api/client.ts
@@ -357,13 +357,23 @@ pollingApiClient.interceptors.response.use(
     }
     // Check if it's a subscription-related error and handle it globally
     if (error.response?.status === 429 || error.response?.status === 402) {
-      console.log('Polling API Client: Detected subscription error, triggering global handler');
+      console.log('Polling API Client: Detected subscription error, triggering global handler', {
+        status: error.response?.status,
+        data: error.response?.data,
+        hasHandler: !!globalSubscriptionErrorHandler
+      });
       if (globalSubscriptionErrorHandler) {
         const wasHandled = globalSubscriptionErrorHandler(error);
+        console.log('Polling API Client: Global handler returned', wasHandled);
         if (wasHandled) {
-          console.log('Polling API Client: Subscription error handled by global handler');
-          return Promise.reject(error);
+          console.log('Polling API Client: Subscription error handled by global handler - modal should be showing');
+        } else {
+          console.warn('Polling API Client: Global handler did not handle subscription error');
         }
+        // Always reject so the polling hook can also handle it
+        return Promise.reject(error);
+      } else {
+        console.warn('Polling API Client: No global subscription error handler registered');
       }
     }
 
diff --git a/frontend/src/components/BlogWriter/BlogWriter.tsx b/frontend/src/components/BlogWriter/BlogWriter.tsx
index bf68d89d..aa5393aa 100644
--- a/frontend/src/components/BlogWriter/BlogWriter.tsx
+++ b/frontend/src/components/BlogWriter/BlogWriter.tsx
@@ -1,49 +1,39 @@
-import React, { useState, useEffect, useRef, useCallback } from 'react';
+import React, { useRef, useCallback } from 'react';
 import { debug } from '../../utils/debug';
-import { CopilotSidebar } from '@copilotkit/react-ui';
-import { useCopilotChatHeadless_c } from '@copilotkit/react-core';
-import { useCopilotAction } from '@copilotkit/react-core';
-import '@copilotkit/react-ui/styles.css';
 import WriterCopilotSidebar from './BlogWriterUtils/WriterCopilotSidebar';
-import { blogWriterApi, BlogSEOActionableRecommendation } from '../../services/blogWriterApi';
-import { useOutlinePolling, useMediumGenerationPolling, useResearchPolling, useRewritePolling } from '../../hooks/usePolling';
+import { blogWriterApi } from '../../services/blogWriterApi';
 import { useClaimFixer } from '../../hooks/useClaimFixer';
 import { useMarkdownProcessor } from '../../hooks/useMarkdownProcessor';
 import { useBlogWriterState } from '../../hooks/useBlogWriterState';
-import { useSuggestions } from './SuggestionsGenerator';
-import EnhancedOutlineEditor from './EnhancedOutlineEditor';
-import ContinuityBadge from './ContinuityBadge';
-import EnhancedTitleSelector from './EnhancedTitleSelector';
-import SEOMiniPanel from './SEOMiniPanel';
-import ResearchResults from './ResearchResults';
-import KeywordInputForm from './KeywordInputForm';
-import ResearchAction from './ResearchAction';
-import { CustomOutlineForm } from './CustomOutlineForm';
-import { ResearchDataActions } from './ResearchDataActions';
-import { EnhancedOutlineActions } from './EnhancedOutlineActions';
 import HallucinationChecker from './HallucinationChecker';
-import { RewriteFeedbackForm } from './RewriteFeedbackForm';
 import Publisher from './Publisher';
 import OutlineGenerator from './OutlineGenerator';
 import OutlineRefiner from './OutlineRefiner';
 import { SEOProcessor } from './SEO';
-import BlogWriterLanding from './BlogWriterLanding';
-import { OutlineProgressModal } from './OutlineProgressModal';
 import TaskProgressModals from './BlogWriterUtils/TaskProgressModals';
-import OutlineFeedbackForm from './OutlineFeedbackForm';
-import { BlogEditor } from './WYSIWYG';
 import { SEOAnalysisModal } from './SEOAnalysisModal';
 import { SEOMetadataModal } from './SEOMetadataModal';
-import PhaseNavigation from './PhaseNavigation';
 import { usePhaseNavigation } from '../../hooks/usePhaseNavigation';
 import HeaderBar from './BlogWriterUtils/HeaderBar';
 import PhaseContent from './BlogWriterUtils/PhaseContent';
 import useBlogWriterCopilotActions from './BlogWriterUtils/useBlogWriterCopilotActions';
-
-// Type assertion for CopilotKit action
-const useCopilotActionTyped = useCopilotAction as any;
+import { useCopilotKitHealth } from '../../hooks/useCopilotKitHealth';
+import { useSEOManager } from './BlogWriterUtils/useSEOManager';
+import { usePhaseActionHandlers } from './BlogWriterUtils/usePhaseActionHandlers';
+import { useBlogWriterPolling } from './BlogWriterUtils/useBlogWriterPolling';
+import { useCopilotSuggestions } from './BlogWriterUtils/useCopilotSuggestions';
+import { usePhaseRestoration } from './BlogWriterUtils/usePhaseRestoration';
+import { useModalVisibility } from './BlogWriterUtils/useModalVisibility';
+import { useBlogWriterRefs } from './BlogWriterUtils/useBlogWriterRefs';
+import { BlogWriterLandingSection } from './BlogWriterUtils/BlogWriterLandingSection';
+import { CopilotKitComponents } from './BlogWriterUtils/CopilotKitComponents';
 
 export const BlogWriter: React.FC = () => {
+  // Check CopilotKit health status
+  const { isAvailable: copilotKitAvailable } = useCopilotKitHealth({
+    enabled: true, // Enable health checking
+  });
+
   // Use custom hook for all state management
   const {
     research,
@@ -91,17 +81,64 @@ export const BlogWriter: React.FC = () => {
     handleContentSave
   } = useBlogWriterState();
 
-  const [isSEOAnalysisModalOpen, setIsSEOAnalysisModalOpen] = useState(false);
-  const [isSEOMetadataModalOpen, setIsSEOMetadataModalOpen] = useState(false);
-  const [seoRecommendationsApplied, setSeoRecommendationsApplied] = useState(false);
-  const lastSEOModalOpenRef = useRef<number>(0);
+  // SEO Manager - handles all SEO-related logic
+  // Initialize phase navigation with temporary false value for seoRecommendationsApplied
+  const [tempSeoRecommendationsApplied] = React.useState(false);
+  const {
+    phases: tempPhases,
+    currentPhase: tempCurrentPhase,
+    navigateToPhase: tempNavigateToPhase,
+    setCurrentPhase: tempSetCurrentPhase,
+    resetUserSelection
+  } = usePhaseNavigation(
+    research,
+    outline,
+    outlineConfirmed,
+    Object.keys(sections).length > 0,
+    contentConfirmed,
+    seoAnalysis,
+    seoMetadata,
+    tempSeoRecommendationsApplied
+  );
 
-  // Phase navigation hook
+  const {
+    isSEOAnalysisModalOpen,
+    setIsSEOAnalysisModalOpen,
+    isSEOMetadataModalOpen,
+    setIsSEOMetadataModalOpen,
+    seoRecommendationsApplied,
+    setSeoRecommendationsApplied,
+    lastSEOModalOpenRef,
+    runSEOAnalysisDirect,
+    handleApplySeoRecommendations,
+    handleSEOAnalysisComplete,
+    handleSEOModalClose,
+    confirmBlogContent,
+  } = useSEOManager({
+    sections,
+    research,
+    outline,
+    selectedTitle,
+    contentConfirmed,
+    seoAnalysis,
+    currentPhase: tempCurrentPhase,
+    navigateToPhase: tempNavigateToPhase,
+    setContentConfirmed,
+    setSeoAnalysis,
+    setSeoMetadata,
+    setSections,
+    setSelectedTitle: setSelectedTitle as (title: string | null) => void,
+    setContinuityRefresh,
+    setFlowAnalysisCompleted,
+    setFlowAnalysisResults,
+  });
+
+  // Phase navigation hook with correct seoRecommendationsApplied
   const {
     phases,
     currentPhase,
     navigateToPhase,
-    resetUserSelection
+    setCurrentPhase,
   } = usePhaseNavigation(
     research,
     outline,
@@ -113,204 +150,17 @@ export const BlogWriter: React.FC = () => {
     seoRecommendationsApplied
   );
 
-  // Helper: run same checks as analyzeSEO and open modal
-  const runSEOAnalysisDirect = (): string => {
-    const hasSections = !!sections && Object.keys(sections).length > 0;
-    const hasResearch = !!research && !!(research as any).keyword_analysis;
-    if (!hasSections) return "No blog content available for SEO analysis. Please generate content first.";
-    if (!hasResearch) return "Research data is required for SEO analysis. Please run research first.";
-    // Prevent rapid re-opens
-    const now = Date.now();
-    if (isSEOAnalysisModalOpen && now - lastSEOModalOpenRef.current < 1000) {
-      return "SEO analysis is already open.";
-    }
-    
-    // Mark content phase as done when user clicks "Next: Run SEO Analysis"
-    if (!contentConfirmed) {
-      setContentConfirmed(true);
-      debug.log('[BlogWriter] Content phase marked as done (SEO analysis triggered)');
-    }
-    
-    setSeoRecommendationsApplied(false);
-    if (!isSEOAnalysisModalOpen) {
-      setIsSEOAnalysisModalOpen(true);
-      lastSEOModalOpenRef.current = now;
-      debug.log('[BlogWriter] SEO modal opened (direct)');
-    }
-    return "Running SEO analysis of your blog content. This will analyze content structure, keyword optimization, readability, and provide actionable recommendations.";
-  };
+  // Phase restoration logic
+  usePhaseRestoration({
+    copilotKitAvailable,
+    research,
+    phases,
+    currentPhase,
+    navigateToPhase,
+    setCurrentPhase,
+  });
 
-  const handleApplySeoRecommendations = useCallback(async (
-    recommendations: BlogSEOActionableRecommendation[]
-  ) => {
-    if (!outline || outline.length === 0) {
-      throw new Error('An outline is required before applying recommendations.');
-    }
-
-    const sectionPayload = outline.map((section) => ({
-      id: section.id,
-      heading: section.heading,
-      content: sections[section.id] ?? '',
-    }));
-
-    const response = await blogWriterApi.applySeoRecommendations({
-      title: selectedTitle || outline[0]?.heading || 'Untitled Blog',
-      sections: sectionPayload,
-      outline,
-      research: (research as any) || {},
-      recommendations,
-    });
-
-    if (!response.success) {
-      throw new Error(response.error || 'Failed to apply recommendations.');
-    }
-
-    if (!response.sections || !Array.isArray(response.sections)) {
-      throw new Error('Recommendation response did not include updated sections.');
-    }
-
-    // Update sections - create new object reference to trigger React re-render
-    const newSections: Record<string, string> = {};
-    response.sections.forEach((section) => {
-      if (section.id && section.content) {
-        newSections[section.id] = section.content;
-      }
-    });
-    
-    // Validate we have sections before updating
-    if (Object.keys(newSections).length === 0) {
-      throw new Error('No valid sections received from SEO recommendations application.');
-    }
-    
-    // Validate sections have actual content
-    const sectionsWithContent = Object.values(newSections).filter(c => c && c.trim().length > 0);
-    if (sectionsWithContent.length === 0) {
-      throw new Error('SEO recommendations resulted in empty sections. Please try again.');
-    }
-    
-    // Log detailed section info for debugging
-    const sectionIds = Object.keys(newSections);
-    const sectionSizes = sectionIds.map(id => ({ id, length: newSections[id]?.length || 0 }));
-    debug.log('[BlogWriter] Applied SEO recommendations: sections updated', { 
-      sectionCount: sectionIds.length,
-      sectionsWithContent: sectionsWithContent.length,
-      sectionIds: sectionIds,
-      sectionSizes: sectionSizes,
-      totalContentLength: Object.values(newSections).reduce((sum, c) => sum + (c?.length || 0), 0)
-    });
-    
-    // Update sections state
-    setSections(newSections);
-    
-    // Force a delay to ensure React processes the state update before proceeding
-    // This gives React time to re-render with new sections before phase navigation checks
-    await new Promise(resolve => setTimeout(resolve, 200));
-    
-    setContinuityRefresh(Date.now());
-    setFlowAnalysisCompleted(false);
-    setFlowAnalysisResults(null);
-
-    if (response.title && response.title !== selectedTitle) {
-      setSelectedTitle(response.title);
-    }
-
-    if (response.applied) {
-      setSeoAnalysis(prev => prev ? { ...prev, applied_recommendations: response.applied } : prev);
-      debug.log('[BlogWriter] SEO analysis state updated with applied recommendations');
-    }
-
-    // Mark recommendations as applied (this will trigger phase navigation check)
-    // But we'll stay in SEO phase to show updated content
-    setSeoRecommendationsApplied(true);
-    debug.log('[BlogWriter] seoRecommendationsApplied set to true');
-    
-    // Ensure we stay in SEO phase to show updated content
-    // Force navigation to SEO phase if we're not already there (safeguard)
-    if (currentPhase !== 'seo') {
-      navigateToPhase('seo');
-      debug.log('[BlogWriter] Forced navigation to SEO phase after applying recommendations');
-    } else {
-      debug.log('[BlogWriter] Already in SEO phase, staying to show updated content');
-    }
-  }, [outline, sections, selectedTitle, research, setSections, setSelectedTitle, setContinuityRefresh, setFlowAnalysisCompleted, setFlowAnalysisResults, setSeoAnalysis, currentPhase, navigateToPhase]);
-
-  // Handle SEO analysis completion
-  const handleSEOAnalysisComplete = useCallback((analysis: any) => {
-    setSeoAnalysis(analysis);
-    debug.log('[BlogWriter] SEO analysis completed', { hasAnalysis: !!analysis });
-  }, [setSeoAnalysis]);
-
-  // Handle SEO modal close - mark SEO phase as done if not already marked
-  const handleSEOModalClose = useCallback(() => {
-    // Mark SEO phase as done when modal closes (even without applying recommendations)
-    if (!seoAnalysis) {
-      // Set a minimal valid seoAnalysis object to mark phase as complete
-      setSeoAnalysis({
-        success: true,
-        overall_score: 0,
-        category_scores: {},
-        analysis_summary: {
-          overall_grade: 'N/A',
-          status: 'Skipped',
-          strongest_category: 'N/A',
-          weakest_category: 'N/A',
-          key_strengths: [],
-          key_weaknesses: [],
-          ai_summary: 'SEO analysis was skipped by user'
-        },
-        actionable_recommendations: [],
-        generated_at: new Date().toISOString()
-      });
-      debug.log('[BlogWriter] SEO phase marked as done (modal closed without analysis)');
-    }
-    setIsSEOAnalysisModalOpen(false);
-    debug.log('[BlogWriter] SEO modal closed');
-  }, [seoAnalysis, setSeoAnalysis, setIsSEOAnalysisModalOpen]);
-
-  // Mark SEO phase as completed when recommendations are applied
-  useEffect(() => {
-    if (seoRecommendationsApplied && seoAnalysis) {
-      // SEO phase is considered complete when recommendations are applied
-      // But stay in SEO phase to show updated content
-      debug.log('[BlogWriter] SEO recommendations applied, SEO phase marked as complete');
-      
-      // Ensure we stay in SEO phase to show updated content (override auto-progression)
-      if (currentPhase !== 'seo' && Object.keys(sections).length > 0) {
-        navigateToPhase('seo');
-        debug.log('[BlogWriter] Forced stay in SEO phase to show updated content');
-      }
-    }
-  }, [seoRecommendationsApplied, seoAnalysis, currentPhase, sections, navigateToPhase]);
-
-  // Track when outlines/content become available for the first time
-  const prevOutlineLenRef = useRef<number>(outline.length);
-  const prevOutlineConfirmedRef = useRef<boolean>(outlineConfirmed);
-  const prevContentConfirmedRef = useRef<boolean>(contentConfirmed);
-  
-  useEffect(() => {
-    const prevLen = prevOutlineLenRef.current;
-    if (research && prevLen === 0 && outline.length > 0) {
-      resetUserSelection();
-    }
-    prevOutlineLenRef.current = outline.length;
-  }, [research, outline.length, resetUserSelection]);
-
-  // Only reset user selection when transitioning from not-confirmed to confirmed
-  useEffect(() => {
-    const wasConfirmed = prevOutlineConfirmedRef.current;
-    if (!wasConfirmed && outlineConfirmed && Object.keys(sections).length > 0) {
-      resetUserSelection(); // Allow auto-progression to content phase
-    }
-    prevOutlineConfirmedRef.current = outlineConfirmed;
-  }, [outlineConfirmed, sections, resetUserSelection]);
-
-  useEffect(() => {
-    const wasConfirmed = prevContentConfirmedRef.current;
-    if (!wasConfirmed && contentConfirmed && seoAnalysis) {
-      resetUserSelection(); // Allow auto-progression to SEO phase
-    }
-    prevContentConfirmedRef.current = contentConfirmed;
-  }, [contentConfirmed, seoAnalysis, resetUserSelection]);
+  // All SEO management logic is now in useSEOManager hook above
 
   // Custom hooks for complex functionality
   const { buildFullMarkdown, buildUpdatedMarkdownForClaim, applyClaimFix } = useClaimFixer(
@@ -324,68 +174,48 @@ export const BlogWriter: React.FC = () => {
     sections
   );
 
-  // Research polling hook (for context awareness)
-  const researchPolling = useResearchPolling({
-    onComplete: handleResearchComplete,
-    onError: (error) => console.error('Research polling error:', error)
+  // Polling hooks - extracted to useBlogWriterPolling
+  const {
+    researchPolling,
+    outlinePolling,
+    mediumPolling,
+    rewritePolling,
+    researchPollingState,
+    outlinePollingState,
+    mediumPollingState,
+  } = useBlogWriterPolling({
+    onResearchComplete: handleResearchComplete,
+    onOutlineComplete: handleOutlineComplete,
+    onOutlineError: handleOutlineError,
+    onSectionsUpdate: setSections,
   });
 
-  // Outline polling hook
-  const outlinePolling = useOutlinePolling({
-    onComplete: handleOutlineComplete,
-    onError: handleOutlineError
+  // Modal visibility management - extracted to useModalVisibility
+  const {
+    showModal,
+    showOutlineModal,
+    setShowOutlineModal,
+    isMediumGenerationStarting,
+    setIsMediumGenerationStarting,
+  } = useModalVisibility({
+    mediumPolling,
+    rewritePolling,
+    outlinePolling,
   });
 
-  // Medium generation polling (used after confirm if short blog)
-  const mediumPolling = useMediumGenerationPolling({
-    onComplete: (result: any) => {
-      try {
-        if (result && result.sections) {
-          const newSections: Record<string, string> = {};
-          result.sections.forEach((s: any) => {
-            newSections[String(s.id)] = s.content || '';
-          });
-          setSections(newSections);
-        }
-      } catch (e) {
-        console.error('Failed to apply medium generation result:', e);
-      }
-    },
-    onError: (err) => console.error('Medium generation failed:', err)
-  });
-
-  // Rewrite polling hook (used for blog rewrite operations)
-  const rewritePolling = useRewritePolling({
-    onComplete: (result: any) => {
-      try {
-        if (result && result.sections) {
-          const newSections: Record<string, string> = {};
-          result.sections.forEach((s: any) => {
-            newSections[String(s.id)] = s.content || '';
-          });
-          setSections(newSections);
-        }
-      } catch (e) {
-        console.error('Failed to apply rewrite result:', e);
-      }
-    },
-    onError: (err) => console.error('Rewrite failed:', err)
-  });
-
-  // Add minimum display time for modal
-  const [showModal, setShowModal] = useState(false);
-  const [modalStartTime, setModalStartTime] = useState<number | null>(null);
-  const [isMediumGenerationStarting, setIsMediumGenerationStarting] = useState(false);
-  const [showOutlineModal, setShowOutlineModal] = useState(false);
-  
-  const suggestions = useSuggestions({
+  // CopilotKit suggestions management - extracted to useCopilotSuggestions
+  const hasContent = React.useMemo(() => Object.keys(sections).length > 0, [sections]);
+  const {
+    suggestions,
+    setSuggestionsRef,
+  } = useCopilotSuggestions({
     research,
     outline,
     outlineConfirmed,
-    researchPolling: { isPolling: researchPolling.isPolling, currentStatus: researchPolling.currentStatus },
-    outlinePolling: { isPolling: outlinePolling.isPolling, currentStatus: outlinePolling.currentStatus },
-    mediumPolling: { isPolling: mediumPolling.isPolling, currentStatus: mediumPolling.currentStatus },
-    hasContent: Object.keys(sections).length > 0,
+    researchPollingState,
+    outlinePollingState,
+    mediumPollingState,
+    hasContent,
     flowAnalysisCompleted,
     contentConfirmed,
     seoAnalysis,
@@ -393,29 +223,17 @@ export const BlogWriter: React.FC = () => {
     seoRecommendationsApplied,
   });
 
-  // Drive CopilotKit suggestions programmatically
-  const copilotHeadless = (useCopilotChatHeadless_c as any)?.();
-  const setSuggestionsRef = useRef<any>(null);
-  useEffect(() => {
-    setSuggestionsRef.current = copilotHeadless?.setSuggestions;
-  }, [copilotHeadless]);
-
-  const suggestionsPayload = React.useMemo(
-    () => (Array.isArray(suggestions) ? suggestions.map((s: any) => ({ title: s.title, message: s.message })) : []),
-    [suggestions]
-  );
-  const prevSuggestionsRef = useRef<string>("__init__");
-  const suggestionsJson = React.useMemo(() => JSON.stringify(suggestionsPayload), [suggestionsPayload]);
-  useEffect(() => {
-    try {
-      if (!setSuggestionsRef.current) return;
-      if (suggestionsJson !== prevSuggestionsRef.current) {
-        setSuggestionsRef.current(suggestionsPayload);
-        debug.log('[BlogWriter] Copilot suggestions pushed', { count: suggestionsPayload.length });
-        prevSuggestionsRef.current = suggestionsJson;
-      }
-    } catch {}
-  }, [suggestionsJson, suggestionsPayload]);
+  // Refs and tracking logic - extracted to useBlogWriterRefs
+  useBlogWriterRefs({
+    research,
+    outline,
+    outlineConfirmed,
+    contentConfirmed,
+    sections,
+    currentPhase,
+    isSEOAnalysisModalOpen,
+    resetUserSelection,
+  });
 
   const handlePhaseClick = useCallback((phaseId: string) => {
     navigateToPhase(phaseId);
@@ -427,40 +245,50 @@ export const BlogWriter: React.FC = () => {
         runSEOAnalysisDirect();
       }
     }
-  }, [navigateToPhase, seoAnalysis, runSEOAnalysisDirect]);
+  }, [navigateToPhase, seoAnalysis, runSEOAnalysisDirect, setIsSEOAnalysisModalOpen]);
+
   const outlineGenRef = useRef<any>(null);
 
-  useEffect(() => {
-    if ((mediumPolling.isPolling || rewritePolling.isPolling || isMediumGenerationStarting) && !showModal) {
-      setShowModal(true);
-      setModalStartTime(Date.now());
-    } else if (!mediumPolling.isPolling && !rewritePolling.isPolling && !isMediumGenerationStarting && showModal) {
-      const elapsed = Date.now() - (modalStartTime || 0);
-      const minDisplayTime = 2000; // 2 seconds minimum
-      
-      if (elapsed < minDisplayTime) {
-        setTimeout(() => {
-          setShowModal(false);
-          setModalStartTime(null);
-        }, minDisplayTime - elapsed);
-      } else {
-        setShowModal(false);
-        setModalStartTime(null);
-      }
+  // Callback to handle cached outline completion
+  const handleCachedOutlineComplete = useCallback((result: { outline: any[], title_options?: string[] }) => {
+    if (result.outline && Array.isArray(result.outline)) {
+      handleOutlineComplete(result);
     }
-  }, [mediumPolling.isPolling, rewritePolling.isPolling, isMediumGenerationStarting, showModal, modalStartTime]);
+  }, [handleOutlineComplete]);
 
-  // Handle outline modal visibility
-  useEffect(() => {
-    if (outlinePolling.isPolling && !showOutlineModal) {
-      setShowOutlineModal(true);
-    } else if (!outlinePolling.isPolling && showOutlineModal) {
-      // Add a small delay to ensure user sees completion message
-      setTimeout(() => {
-        setShowOutlineModal(false);
-      }, 1000);
+  // Callback to handle cached content completion
+  const handleCachedContentComplete = useCallback((cachedSections: Record<string, string>) => {
+    if (cachedSections && Object.keys(cachedSections).length > 0) {
+      setSections(cachedSections);
+      debug.log('[BlogWriter] Cached content loaded into state', { sections: Object.keys(cachedSections).length });
     }
-  }, [outlinePolling.isPolling, showOutlineModal]);
+  }, [setSections]);
+
+  // Phase action handlers for when CopilotKit is unavailable - extracted to usePhaseActionHandlers
+  const {
+    handleResearchAction,
+    handleOutlineAction,
+    handleContentAction,
+    handleSEOAction,
+    handlePublishAction,
+  } = usePhaseActionHandlers({
+    research,
+    outline,
+    selectedTitle,
+    contentConfirmed,
+    sections,
+    navigateToPhase,
+    handleOutlineConfirmed,
+    setIsMediumGenerationStarting,
+    mediumPolling,
+    outlineGenRef,
+    setOutline,
+    setContentConfirmed,
+    setIsSEOMetadataModalOpen,
+    runSEOAnalysisDirect,
+    onOutlineComplete: handleCachedOutlineComplete,
+    onContentComplete: handleCachedContentComplete,
+  });
 
   // Handle medium generation start from OutlineFeedbackForm
   const handleMediumGenerationStarted = (taskId: string) => {
@@ -475,77 +303,11 @@ export const BlogWriter: React.FC = () => {
     setIsMediumGenerationStarting(true);
   };
 
-  // Debug medium polling state
-  console.log('Medium polling state:', {
-    isPolling: mediumPolling.isPolling,
-    status: mediumPolling.currentStatus,
-    progressCount: mediumPolling.progressMessages.length
-  });
-
-  // Log critical state changes only (reduce noise)
-  const lastPhaseRef = useRef<string>('');
-  const lastSeoOpenRef = useRef<boolean>(false);
-  const lastSectionsLenRef = useRef<number>(0);
-
-  useEffect(() => {
-    if (currentPhase !== lastPhaseRef.current) {
-      debug.log('[BlogWriter] Phase changed', { currentPhase });
-      lastPhaseRef.current = currentPhase;
-    }
-  }, [currentPhase]);
-
-  useEffect(() => {
-    const open = isSEOAnalysisModalOpen;
-    if (open !== lastSeoOpenRef.current) {
-      debug.log('[BlogWriter] SEO modal', { isOpen: open });
-      lastSeoOpenRef.current = open;
-    }
-  }, [isSEOAnalysisModalOpen]);
-
-  useEffect(() => {
-    const len = Object.keys(sections || {}).length;
-    if (len !== lastSectionsLenRef.current) {
-      debug.log('[BlogWriter] Sections updated', { count: len });
-      lastSectionsLenRef.current = len;
-    }
-  }, [sections]);
-
-  useEffect(() => {
-    debug.log('[BlogWriter] Suggestions updated', { suggestions });
-  }, [suggestions]);
-
-  // Force-sync Copilot suggestions right after SEO recommendations applied (guarded by previous suggestions key)
-  useEffect(() => {
-    if (!seoAnalysis || !seoRecommendationsApplied || !setSuggestionsRef.current) return;
-    try {
-      if (suggestionsJson !== prevSuggestionsRef.current) {
-        setSuggestionsRef.current(suggestionsPayload);
-        debug.log('[BlogWriter] Forced Copilot suggestions sync after SEO recommendations applied', { count: suggestionsPayload.length });
-        prevSuggestionsRef.current = suggestionsJson;
-      }
-    } catch (e) {
-      console.error('Failed to push Copilot suggestions after SEO apply:', e);
-    }
-  }, [seoAnalysis, seoRecommendationsApplied, suggestionsJson, suggestionsPayload]);
-
-  const confirmBlogContentCb = useCallback(() => {
-    debug.log('[BlogWriter] Blog content confirmed by user');
-    setContentConfirmed(true);
-    resetUserSelection();
-    setSeoRecommendationsApplied(false);
-    navigateToPhase('seo');
-    setTimeout(() => {
-      setIsSEOAnalysisModalOpen(true);
-      debug.log('[BlogWriter] SEO modal opened (confirm→direct)');
-    }, 0);
-    return "✅ Blog content has been confirmed! Running SEO analysis now.";
-  }, [setContentConfirmed, resetUserSelection, navigateToPhase, setIsSEOAnalysisModalOpen]);
-
   useBlogWriterCopilotActions({
     isSEOAnalysisModalOpen,
     lastSEOModalOpenRef,
     runSEOAnalysisDirect,
-    confirmBlogContent: confirmBlogContentCb,
+    confirmBlogContent,
     sections,
     research,
     openSEOMetadata: () => setIsSEOMetadataModalOpen(true),
@@ -557,47 +319,22 @@ export const BlogWriter: React.FC = () => {
 
   return (
     <div style={{ height: '100vh', display: 'flex', flexDirection: 'column' }}>
-      {/* Extracted Components */}
-      <KeywordInputForm 
-        onResearchComplete={handleResearchComplete}
-        onTaskStart={(taskId) => researchPolling.startPolling(taskId)}
-      />
-      <CustomOutlineForm onOutlineCreated={setOutline} />
-      <ResearchAction onResearchComplete={handleResearchComplete} />
-      <ResearchDataActions 
-        research={research} 
-        onOutlineCreated={setOutline} 
-        onTitleOptionsSet={setTitleOptions} 
-      />
-      <EnhancedOutlineActions 
-        outline={outline} 
-        onOutlineUpdated={setOutline} 
-      />
-      <OutlineFeedbackForm 
-        outline={outline} 
-        research={research!} 
-        onOutlineConfirmed={handleOutlineConfirmed}
-        onOutlineRefined={handleOutlineRefined}
-        onMediumGenerationStarted={handleMediumGenerationStarted}
-        onMediumGenerationTriggered={handleMediumGenerationTriggered}
-        sections={sections}
-        blogTitle={selectedTitle}
-        onFlowAnalysisComplete={(analysis) => {
-          console.log('Flow analysis completed:', analysis);
-          setFlowAnalysisCompleted(true);
-          setFlowAnalysisResults(analysis);
-          // Trigger a refresh of continuity badges
-          setContinuityRefresh((prev: number) => (prev || 0) + 1);
-        }}
-      />
-      
-      {/* Rewrite Feedback Form - Only show when content exists */}
-      {Object.keys(sections).length > 0 && (
-        <RewriteFeedbackForm
-          research={research!}
+      {/* CopilotKit-dependent components - extracted to CopilotKitComponents */}
+      {copilotKitAvailable && (
+        <CopilotKitComponents
+          research={research}
           outline={outline}
+          outlineConfirmed={outlineConfirmed}
           sections={sections}
-          blogTitle={selectedTitle}
+          selectedTitle={selectedTitle}
+          onResearchComplete={handleResearchComplete}
+          onOutlineCreated={setOutline}
+          onOutlineUpdated={setOutline}
+          onTitleOptionsSet={setTitleOptions}
+          onOutlineConfirmed={handleOutlineConfirmed}
+          onOutlineRefined={(feedback?: string) => handleOutlineRefined(feedback || '')}
+          onMediumGenerationStarted={handleMediumGenerationStarted}
+          onMediumGenerationTriggered={handleMediumGenerationTriggered}
           onRewriteStarted={(taskId) => {
             console.log('Starting rewrite polling for task:', taskId);
             rewritePolling.startPolling(taskId);
@@ -606,6 +343,10 @@ export const BlogWriter: React.FC = () => {
             console.log('Rewrite triggered - showing modal immediately');
             setIsMediumGenerationStarting(true);
           }}
+          setFlowAnalysisCompleted={setFlowAnalysisCompleted}
+          setFlowAnalysisResults={setFlowAnalysisResults}
+          setContinuityRefresh={setContinuityRefresh}
+          researchPolling={researchPolling}
         />
       )}
       
@@ -638,19 +379,41 @@ export const BlogWriter: React.FC = () => {
         seoMetadata={seoMetadata}
       />
       
-      {!research ? (
-        <BlogWriterLanding 
-          onStartWriting={() => {
-            // Trigger the copilot to start the research process
+      {/* Always show HeaderBar when CopilotKit is unavailable, or when research exists */}
+      {(!copilotKitAvailable || research) && (
+        <HeaderBar
+          phases={phases}
+          currentPhase={currentPhase}
+          onPhaseClick={handlePhaseClick}
+          copilotKitAvailable={copilotKitAvailable}
+          actionHandlers={{
+            onResearchAction: handleResearchAction,
+            onOutlineAction: handleOutlineAction,
+            onContentAction: handleContentAction,
+            onSEOAction: handleSEOAction,
+            onPublishAction: handlePublishAction,
           }}
+          hasResearch={!!research}
+          hasOutline={outline.length > 0}
+          outlineConfirmed={outlineConfirmed}
+          hasContent={Object.keys(sections).length > 0}
+          contentConfirmed={contentConfirmed}
+          hasSEOAnalysis={!!seoAnalysis}
+          hasSEOMetadata={!!seoMetadata}
         />
-      ) : (
-        <>
-      <HeaderBar
-        phases={phases}
+      )}
+
+      {/* Landing section - extracted to BlogWriterLandingSection */}
+      <BlogWriterLandingSection
+        research={research}
+        copilotKitAvailable={copilotKitAvailable}
         currentPhase={currentPhase}
-        onPhaseClick={handlePhaseClick}
+        navigateToPhase={navigateToPhase}
+        onResearchComplete={handleResearchComplete}
       />
+
+      {research && (
+        <>
       <PhaseContent
         currentPhase={currentPhase}
         research={research}
@@ -679,6 +442,14 @@ export const BlogWriter: React.FC = () => {
         seoMetadata={seoMetadata}
         onTitleSelect={handleTitleSelect}
         onCustomTitle={handleCustomTitle}
+        copilotKitAvailable={copilotKitAvailable}
+        onResearchComplete={handleResearchComplete}
+        onOutlineGenerationStart={(taskId) => {
+          setOutlineTaskId(taskId);
+          outlinePolling.startPolling(taskId);
+          setShowOutlineModal(true);
+        }}
+        onContentGenerationStart={handleMediumGenerationStarted}
       />
         </>
       )}
diff --git a/frontend/src/components/BlogWriter/BlogWriterUtils/BlogWriterLandingSection.tsx b/frontend/src/components/BlogWriter/BlogWriterUtils/BlogWriterLandingSection.tsx
new file mode 100644
index 00000000..f05b19c6
--- /dev/null
+++ b/frontend/src/components/BlogWriter/BlogWriterUtils/BlogWriterLandingSection.tsx
@@ -0,0 +1,46 @@
+import React from 'react';
+import BlogWriterLanding from '../BlogWriterLanding';
+import ManualResearchForm from '../ManualResearchForm';
+
+interface BlogWriterLandingSectionProps {
+  research: any;
+  copilotKitAvailable: boolean;
+  currentPhase: string;
+  navigateToPhase: (phase: string) => void;
+  onResearchComplete: (research: any) => void;
+}
+
+export const BlogWriterLandingSection: React.FC<BlogWriterLandingSectionProps> = ({
+  research,
+  copilotKitAvailable,
+  currentPhase,
+  navigateToPhase,
+  onResearchComplete,
+}) => {
+  if (!research) {
+    return (
+      <>
+        {!copilotKitAvailable && currentPhase === 'research' && (
+          <ManualResearchForm onResearchComplete={onResearchComplete} />
+        )}
+        {copilotKitAvailable && (
+          <BlogWriterLanding 
+            onStartWriting={() => {
+              // Trigger the copilot to start the research process
+            }}
+          />
+        )}
+        {!copilotKitAvailable && currentPhase !== 'research' && (
+          <BlogWriterLanding 
+            onStartWriting={() => {
+              // Navigate to research phase when CopilotKit unavailable
+              navigateToPhase('research');
+            }}
+          />
+        )}
+      </>
+    );
+  }
+  return null;
+};
+
diff --git a/frontend/src/components/BlogWriter/BlogWriterUtils/CopilotKitComponents.tsx b/frontend/src/components/BlogWriter/BlogWriterUtils/CopilotKitComponents.tsx
new file mode 100644
index 00000000..7b708e85
--- /dev/null
+++ b/frontend/src/components/BlogWriter/BlogWriterUtils/CopilotKitComponents.tsx
@@ -0,0 +1,103 @@
+import React from 'react';
+import KeywordInputForm from '../KeywordInputForm';
+import ResearchAction from '../ResearchAction';
+import { CustomOutlineForm } from '../CustomOutlineForm';
+import { ResearchDataActions } from '../ResearchDataActions';
+import { EnhancedOutlineActions } from '../EnhancedOutlineActions';
+import OutlineFeedbackForm from '../OutlineFeedbackForm';
+import { RewriteFeedbackForm } from '../RewriteFeedbackForm';
+
+interface CopilotKitComponentsProps {
+  research: any;
+  outline: any[];
+  outlineConfirmed: boolean;
+  sections: Record<string, string>;
+  selectedTitle: string | null;
+  onResearchComplete: (research: any) => void;
+  onOutlineCreated: (outline: any[]) => void;
+  onOutlineUpdated: (outline: any[]) => void;
+  onTitleOptionsSet: (titles: any[]) => void;
+  onOutlineConfirmed: () => void;
+  onOutlineRefined: (feedback?: string) => void;
+  onMediumGenerationStarted: (taskId: string) => void;
+  onMediumGenerationTriggered: () => void;
+  onRewriteStarted: (taskId: string) => void;
+  onRewriteTriggered: () => void;
+  setFlowAnalysisCompleted: (completed: boolean) => void;
+  setFlowAnalysisResults: (results: any) => void;
+  setContinuityRefresh: (refresh: number | ((prev: number) => number)) => void;
+  researchPolling: any;
+}
+
+export const CopilotKitComponents: React.FC<CopilotKitComponentsProps> = ({
+  research,
+  outline,
+  outlineConfirmed,
+  sections,
+  selectedTitle,
+  onResearchComplete,
+  onOutlineCreated,
+  onOutlineUpdated,
+  onTitleOptionsSet,
+  onOutlineConfirmed,
+  onOutlineRefined,
+  onMediumGenerationStarted,
+  onMediumGenerationTriggered,
+  onRewriteStarted,
+  onRewriteTriggered,
+  setFlowAnalysisCompleted,
+  setFlowAnalysisResults,
+  setContinuityRefresh,
+  researchPolling,
+}) => {
+  return (
+    <>
+      <KeywordInputForm 
+        onResearchComplete={onResearchComplete}
+        onTaskStart={(taskId) => researchPolling.startPolling(taskId)}
+      />
+      <CustomOutlineForm onOutlineCreated={onOutlineCreated} />
+      <ResearchAction onResearchComplete={onResearchComplete} />
+      
+      <ResearchDataActions 
+        research={research} 
+        onOutlineCreated={onOutlineCreated} 
+        onTitleOptionsSet={onTitleOptionsSet} 
+      />
+      <EnhancedOutlineActions 
+        outline={outline} 
+        onOutlineUpdated={onOutlineUpdated} 
+      />
+      <OutlineFeedbackForm
+        outline={outline} 
+        research={research!} 
+        onOutlineConfirmed={onOutlineConfirmed}
+        onOutlineRefined={onOutlineRefined}
+        onMediumGenerationStarted={onMediumGenerationStarted}
+        onMediumGenerationTriggered={onMediumGenerationTriggered}
+        sections={sections}
+        blogTitle={selectedTitle ?? undefined}
+        onFlowAnalysisComplete={(analysis) => {
+          console.log('Flow analysis completed:', analysis);
+          setFlowAnalysisCompleted(true);
+          setFlowAnalysisResults(analysis);
+          // Trigger a refresh of continuity badges
+          setContinuityRefresh((prev: number) => (prev || 0) + 1);
+        }}
+      />
+      
+      {/* Rewrite Feedback Form - Only show when content exists */}
+      {Object.keys(sections).length > 0 && (
+        <RewriteFeedbackForm
+          research={research!}
+          outline={outline}
+          sections={sections}
+          blogTitle={selectedTitle || 'Untitled'}
+          onRewriteStarted={onRewriteStarted}
+          onRewriteTriggered={onRewriteTriggered}
+        />
+      )}
+    </>
+  );
+};
+
diff --git a/frontend/src/components/BlogWriter/BlogWriterUtils/HeaderBar.tsx b/frontend/src/components/BlogWriter/BlogWriterUtils/HeaderBar.tsx
index 14fbe899..bd164cdb 100644
--- a/frontend/src/components/BlogWriter/BlogWriterUtils/HeaderBar.tsx
+++ b/frontend/src/components/BlogWriter/BlogWriterUtils/HeaderBar.tsx
@@ -1,13 +1,35 @@
 import React from 'react';
-import PhaseNavigation from '../PhaseNavigation';
+import PhaseNavigation, { PhaseActionHandlers } from '../PhaseNavigation';
 
 interface HeaderBarProps {
   phases: any[];
   currentPhase: string;
   onPhaseClick: (phaseId: string) => void;
+  copilotKitAvailable?: boolean;
+  actionHandlers?: PhaseActionHandlers;
+  hasResearch?: boolean;
+  hasOutline?: boolean;
+  outlineConfirmed?: boolean;
+  hasContent?: boolean;
+  contentConfirmed?: boolean;
+  hasSEOAnalysis?: boolean;
+  hasSEOMetadata?: boolean;
 }
 
-export const HeaderBar: React.FC<HeaderBarProps> = ({ phases, currentPhase, onPhaseClick }) => {
+export const HeaderBar: React.FC<HeaderBarProps> = ({ 
+  phases, 
+  currentPhase, 
+  onPhaseClick,
+  copilotKitAvailable = true,
+  actionHandlers,
+  hasResearch = false,
+  hasOutline = false,
+  outlineConfirmed = false,
+  hasContent = false,
+  contentConfirmed = false,
+  hasSEOAnalysis = false,
+  hasSEOMetadata = false,
+}) => {
   return (
     <div style={{ padding: 16, borderBottom: '1px solid #eee' }}>
       <div style={{ display: 'flex', justifyContent: 'space-between', alignItems: 'center', marginBottom: 12 }}>
@@ -31,6 +53,15 @@ export const HeaderBar: React.FC<HeaderBarProps> = ({ phases, currentPhase, onPh
         phases={phases}
         currentPhase={currentPhase}
         onPhaseClick={onPhaseClick}
+        copilotKitAvailable={copilotKitAvailable}
+        actionHandlers={actionHandlers}
+        hasResearch={hasResearch}
+        hasOutline={hasOutline}
+        outlineConfirmed={outlineConfirmed}
+        hasContent={hasContent}
+        contentConfirmed={contentConfirmed}
+        hasSEOAnalysis={hasSEOAnalysis}
+        hasSEOMetadata={hasSEOMetadata}
       />
     </div>
   );
diff --git a/frontend/src/components/BlogWriter/BlogWriterUtils/PhaseContent.tsx b/frontend/src/components/BlogWriter/BlogWriterUtils/PhaseContent.tsx
index 71b3c126..eea4ab29 100644
--- a/frontend/src/components/BlogWriter/BlogWriterUtils/PhaseContent.tsx
+++ b/frontend/src/components/BlogWriter/BlogWriterUtils/PhaseContent.tsx
@@ -4,6 +4,9 @@ import EnhancedTitleSelector from '../EnhancedTitleSelector';
 import EnhancedOutlineEditor from '../EnhancedOutlineEditor';
 import { BlogEditor } from '../WYSIWYG';
 import OutlineCtaBanner from './OutlineCtaBanner';
+import ManualResearchForm from '../ManualResearchForm';
+import ManualOutlineButton from '../ManualOutlineButton';
+import ManualContentButton from '../ManualContentButton';
 
 interface PhaseContentProps {
   currentPhase: string;
@@ -33,6 +36,10 @@ interface PhaseContentProps {
   onCustomTitle: any;
   sectionImages?: Record<string, string>;
   setSectionImages?: (images: Record<string, string> | ((prev: Record<string, string>) => Record<string, string>)) => void;
+  copilotKitAvailable?: boolean; // Whether CopilotKit is available
+  onResearchComplete?: (research: any) => void; // Callback when research completes (for manual form)
+  onOutlineGenerationStart?: (taskId: string) => void; // Callback when outline generation starts
+  onContentGenerationStart?: (taskId: string) => void; // Callback when content generation starts
 }
 
 export const PhaseContent: React.FC<PhaseContentProps> = ({
@@ -62,7 +69,11 @@ export const PhaseContent: React.FC<PhaseContentProps> = ({
   onTitleSelect,
   onCustomTitle,
   sectionImages,
-  setSectionImages
+  setSectionImages,
+  copilotKitAvailable = true,
+  onResearchComplete,
+  onOutlineGenerationStart,
+  onContentGenerationStart,
 }) => {
   return (
     <div style={{ display: 'flex', flex: 1, overflow: 'hidden' }}>
@@ -72,10 +83,16 @@ export const PhaseContent: React.FC<PhaseContentProps> = ({
             {research ? (
               <ResearchResults research={research} />
             ) : (
-              <div style={{ padding: '20px', textAlign: 'center' }}>
-                <h3>Start Your Research</h3>
-                <p>Use the copilot to begin researching your blog topic.</p>
-              </div>
+              <>
+                {copilotKitAvailable ? (
+                  <div style={{ padding: '20px', textAlign: 'center' }}>
+                    <h3>Start Your Research</h3>
+                    <p>Use the copilot to begin researching your blog topic.</p>
+                  </div>
+                ) : (
+                  <ManualResearchForm onResearchComplete={onResearchComplete} />
+                )}
+              </>
             )}
           </>
         )}
@@ -83,7 +100,17 @@ export const PhaseContent: React.FC<PhaseContentProps> = ({
         {currentPhase === 'outline' && research && (
           <>
             {outline.length === 0 && (
-              <OutlineCtaBanner onGenerate={() => outlineGenRef.current?.generateNow()} />
+              <>
+                {copilotKitAvailable ? (
+                  <OutlineCtaBanner onGenerate={() => outlineGenRef.current?.generateNow()} />
+                ) : (
+                  <ManualOutlineButton
+                    outlineGenRef={outlineGenRef}
+                    hasResearch={!!research}
+                    onGenerationStart={onOutlineGenerationStart}
+                  />
+                )}
+              </>
             )}
             {outline.length > 0 ? (
               <>
@@ -108,6 +135,12 @@ export const PhaseContent: React.FC<PhaseContentProps> = ({
                   setSectionImages={setSectionImages}
                 />
               </>
+            ) : !copilotKitAvailable ? (
+              <ManualOutlineButton
+                outlineGenRef={outlineGenRef}
+                hasResearch={!!research}
+                onGenerationStart={onOutlineGenerationStart}
+              />
             ) : (
               <div style={{ padding: '20px', textAlign: 'center' }}>
                 <h3>Create Your Outline</h3>
@@ -135,10 +168,22 @@ export const PhaseContent: React.FC<PhaseContentProps> = ({
                 sectionImages={sectionImages}
               />
             ) : (
-              <div style={{ padding: '20px', textAlign: 'center' }}>
-                <h3>Confirm Your Outline</h3>
-                <p>Review and confirm your outline before generating content.</p>
-              </div>
+              <>
+                {copilotKitAvailable ? (
+                  <div style={{ padding: '20px', textAlign: 'center' }}>
+                    <h3>Confirm Your Outline</h3>
+                    <p>Review and confirm your outline before generating content.</p>
+                  </div>
+                ) : (
+                  <ManualContentButton
+                    outline={outline}
+                    research={research}
+                    blogTitle={selectedTitle || undefined}
+                    sections={sections}
+                    onGenerationStart={onContentGenerationStart}
+                  />
+                )}
+              </>
             )}
           </>
         )}
diff --git a/frontend/src/components/BlogWriter/BlogWriterUtils/WixConnectModal.tsx b/frontend/src/components/BlogWriter/BlogWriterUtils/WixConnectModal.tsx
index adf7c225..8c58d97d 100644
--- a/frontend/src/components/BlogWriter/BlogWriterUtils/WixConnectModal.tsx
+++ b/frontend/src/components/BlogWriter/BlogWriterUtils/WixConnectModal.tsx
@@ -81,6 +81,16 @@ export const WixConnectModal: React.FC<WixConnectModalProps> = ({
     try {
       setIsConnecting(true);
       setError(null);
+      // Store current page URL so we can redirect back after OAuth completes
+      // This MUST be stored before calling handleConnect to ensure it's available after redirect
+      // We ALWAYS override any existing redirect URL since we know the exact page we're on (Blog Writer)
+      const currentUrl = window.location.href;
+      try {
+        sessionStorage.setItem('wix_oauth_redirect', currentUrl);
+        console.log('[WixConnectModal] Stored redirect URL (overriding any existing):', currentUrl);
+      } catch (e) {
+        console.warn('[WixConnectModal] Failed to store redirect URL:', e);
+      }
       await handleConnect('wix');
       // OAuth will redirect, so we don't need to do anything else here
       // The postMessage handler or URL param handler will close the modal
diff --git a/frontend/src/components/BlogWriter/BlogWriterUtils/useBlogWriterPolling.ts b/frontend/src/components/BlogWriter/BlogWriterUtils/useBlogWriterPolling.ts
new file mode 100644
index 00000000..d49de00c
--- /dev/null
+++ b/frontend/src/components/BlogWriter/BlogWriterUtils/useBlogWriterPolling.ts
@@ -0,0 +1,101 @@
+import React from 'react';
+import {
+  useResearchPolling,
+  useOutlinePolling,
+  useMediumGenerationPolling,
+  useRewritePolling,
+} from '../../../hooks/usePolling';
+import { blogWriterCache } from '../../../services/blogWriterCache';
+
+interface UseBlogWriterPollingProps {
+  onResearchComplete: (research: any) => void;
+  onOutlineComplete: (outline: any) => void;
+  onOutlineError: (error: any) => void;
+  onSectionsUpdate: (sections: Record<string, string>) => void;
+}
+
+export const useBlogWriterPolling = ({
+  onResearchComplete,
+  onOutlineComplete,
+  onOutlineError,
+  onSectionsUpdate,
+}: UseBlogWriterPollingProps) => {
+  // Research polling hook (for context awareness)
+  const researchPolling = useResearchPolling({
+    onComplete: onResearchComplete,
+    onError: (error) => console.error('Research polling error:', error)
+  });
+
+  // Outline polling hook
+  const outlinePolling = useOutlinePolling({
+    onComplete: onOutlineComplete,
+    onError: onOutlineError
+  });
+
+  // Medium generation polling (used after confirm if short blog)
+  const mediumPolling = useMediumGenerationPolling({
+    onComplete: (result: any) => {
+      try {
+        if (result && result.sections) {
+          const newSections: Record<string, string> = {};
+          result.sections.forEach((s: any) => {
+            newSections[String(s.id)] = s.content || '';
+          });
+          onSectionsUpdate(newSections);
+          
+          // Cache the generated content (shared utility)
+          if (Object.keys(newSections).length > 0) {
+            const sectionIds = Object.keys(newSections);
+            blogWriterCache.cacheContent(newSections, sectionIds);
+          }
+        }
+      } catch (e) {
+        console.error('Failed to apply medium generation result:', e);
+      }
+    },
+    onError: (err) => console.error('Medium generation failed:', err)
+  });
+
+  // Rewrite polling hook (used for blog rewrite operations)
+  const rewritePolling = useRewritePolling({
+    onComplete: (result: any) => {
+      try {
+        if (result && result.sections) {
+          const newSections: Record<string, string> = {};
+          result.sections.forEach((s: any) => {
+            newSections[String(s.id)] = s.content || '';
+          });
+          onSectionsUpdate(newSections);
+        }
+      } catch (e) {
+        console.error('Failed to apply rewrite result:', e);
+      }
+    },
+    onError: (err) => console.error('Rewrite failed:', err)
+  });
+
+  // Memoize polling state objects to prevent unnecessary recalculations
+  const researchPollingState = React.useMemo(
+    () => ({ isPolling: researchPolling.isPolling, currentStatus: researchPolling.currentStatus }),
+    [researchPolling.isPolling, researchPolling.currentStatus]
+  );
+  const outlinePollingState = React.useMemo(
+    () => ({ isPolling: outlinePolling.isPolling, currentStatus: outlinePolling.currentStatus }),
+    [outlinePolling.isPolling, outlinePolling.currentStatus]
+  );
+  const mediumPollingState = React.useMemo(
+    () => ({ isPolling: mediumPolling.isPolling, currentStatus: mediumPolling.currentStatus }),
+    [mediumPolling.isPolling, mediumPolling.currentStatus]
+  );
+
+  return {
+    researchPolling,
+    outlinePolling,
+    mediumPolling,
+    rewritePolling,
+    researchPollingState,
+    outlinePollingState,
+    mediumPollingState,
+  };
+};
+
diff --git a/frontend/src/components/BlogWriter/BlogWriterUtils/useBlogWriterRefs.ts b/frontend/src/components/BlogWriter/BlogWriterUtils/useBlogWriterRefs.ts
new file mode 100644
index 00000000..c28abb73
--- /dev/null
+++ b/frontend/src/components/BlogWriter/BlogWriterUtils/useBlogWriterRefs.ts
@@ -0,0 +1,83 @@
+import { useRef, useEffect } from 'react';
+import { debug } from '../../../utils/debug';
+
+interface UseBlogWriterRefsProps {
+  research: any;
+  outline: any[];
+  outlineConfirmed: boolean;
+  contentConfirmed: boolean;
+  sections: Record<string, string>;
+  currentPhase: string;
+  isSEOAnalysisModalOpen: boolean;
+  resetUserSelection: () => void;
+}
+
+export const useBlogWriterRefs = ({
+  research,
+  outline,
+  outlineConfirmed,
+  contentConfirmed,
+  sections,
+  currentPhase,
+  isSEOAnalysisModalOpen,
+  resetUserSelection,
+}: UseBlogWriterRefsProps) => {
+  // Track when outlines/content become available for the first time
+  const prevOutlineLenRef = useRef<number>(outline.length);
+  const prevOutlineConfirmedRef = useRef<boolean>(outlineConfirmed);
+  const prevContentConfirmedRef = useRef<boolean>(contentConfirmed);
+  
+  useEffect(() => {
+    const prevLen = prevOutlineLenRef.current;
+    if (research && prevLen === 0 && outline.length > 0) {
+      resetUserSelection();
+    }
+    prevOutlineLenRef.current = outline.length;
+  }, [research, outline.length, resetUserSelection]);
+
+  // Only reset user selection when transitioning from not-confirmed to confirmed
+  useEffect(() => {
+    const wasConfirmed = prevOutlineConfirmedRef.current;
+    if (!wasConfirmed && outlineConfirmed && Object.keys(sections).length > 0) {
+      resetUserSelection(); // Allow auto-progression to content phase
+    }
+    prevOutlineConfirmedRef.current = outlineConfirmed;
+  }, [outlineConfirmed, sections, resetUserSelection]);
+
+  useEffect(() => {
+    const wasConfirmed = prevContentConfirmedRef.current;
+    if (!wasConfirmed && contentConfirmed) {
+      resetUserSelection(); // Allow auto-progression to SEO phase
+    }
+    prevContentConfirmedRef.current = contentConfirmed;
+  }, [contentConfirmed, resetUserSelection]);
+
+  // Log critical state changes only (reduce noise)
+  const lastPhaseRef = useRef<string>('');
+  const lastSeoOpenRef = useRef<boolean>(false);
+  const lastSectionsLenRef = useRef<number>(0);
+
+  useEffect(() => {
+    if (currentPhase !== lastPhaseRef.current) {
+      debug.log('[BlogWriter] Phase changed', { currentPhase });
+      lastPhaseRef.current = currentPhase;
+    }
+  }, [currentPhase]);
+
+  useEffect(() => {
+    const open = isSEOAnalysisModalOpen;
+    if (open !== lastSeoOpenRef.current) {
+      debug.log('[BlogWriter] SEO modal', { isOpen: open });
+      lastSeoOpenRef.current = open;
+    }
+  }, [isSEOAnalysisModalOpen]);
+
+  useEffect(() => {
+    const len = Object.keys(sections || {}).length;
+    if (len !== lastSectionsLenRef.current) {
+      debug.log('[BlogWriter] Sections updated', { count: len });
+      lastSectionsLenRef.current = len;
+    }
+  }, [sections]);
+};
+
diff --git a/frontend/src/components/BlogWriter/BlogWriterUtils/useCopilotSuggestions.ts b/frontend/src/components/BlogWriter/BlogWriterUtils/useCopilotSuggestions.ts
new file mode 100644
index 00000000..29c966cd
--- /dev/null
+++ b/frontend/src/components/BlogWriter/BlogWriterUtils/useCopilotSuggestions.ts
@@ -0,0 +1,94 @@
+import React, { useRef, useEffect, useMemo } from 'react';
+import { useCopilotChatHeadless_c } from '@copilotkit/react-core';
+import { debug } from '../../../utils/debug';
+import { useSuggestions } from '../SuggestionsGenerator';
+
+interface UseCopilotSuggestionsProps {
+  research: any;
+  outline: any[];
+  outlineConfirmed: boolean;
+  researchPollingState: { isPolling: boolean; currentStatus: any };
+  outlinePollingState: { isPolling: boolean; currentStatus: any };
+  mediumPollingState: { isPolling: boolean; currentStatus: any };
+  hasContent: boolean;
+  flowAnalysisCompleted: boolean;
+  contentConfirmed: boolean;
+  seoAnalysis: any;
+  seoMetadata: any;
+  seoRecommendationsApplied: boolean;
+}
+
+export const useCopilotSuggestions = ({
+  research,
+  outline,
+  outlineConfirmed,
+  researchPollingState,
+  outlinePollingState,
+  mediumPollingState,
+  hasContent,
+  flowAnalysisCompleted,
+  contentConfirmed,
+  seoAnalysis,
+  seoMetadata,
+  seoRecommendationsApplied,
+}: UseCopilotSuggestionsProps) => {
+  const suggestions = useSuggestions({
+    research,
+    outline,
+    outlineConfirmed,
+    researchPolling: researchPollingState,
+    outlinePolling: outlinePollingState,
+    mediumPolling: mediumPollingState,
+    hasContent,
+    flowAnalysisCompleted,
+    contentConfirmed,
+    seoAnalysis,
+    seoMetadata,
+    seoRecommendationsApplied,
+  });
+
+  // Drive CopilotKit suggestions programmatically
+  const copilotHeadless = (useCopilotChatHeadless_c as any)?.();
+  const setSuggestionsRef = useRef<any>(null);
+  useEffect(() => {
+    setSuggestionsRef.current = copilotHeadless?.setSuggestions;
+  }, [copilotHeadless]);
+
+  const suggestionsPayload = useMemo(
+    () => (Array.isArray(suggestions) ? suggestions.map((s: any) => ({ title: s.title, message: s.message })) : []),
+    [suggestions]
+  );
+  const prevSuggestionsRef = useRef<string>("__init__");
+  const suggestionsJson = useMemo(() => JSON.stringify(suggestionsPayload), [suggestionsPayload]);
+  
+  useEffect(() => {
+    try {
+      if (!setSuggestionsRef.current) return;
+      if (suggestionsJson !== prevSuggestionsRef.current) {
+        setSuggestionsRef.current(suggestionsPayload);
+        debug.log('[BlogWriter] Copilot suggestions pushed', { count: suggestionsPayload.length });
+        prevSuggestionsRef.current = suggestionsJson;
+      }
+    } catch {}
+  }, [suggestionsJson, suggestionsPayload]);
+
+  // Force-sync Copilot suggestions right after SEO recommendations applied
+  useEffect(() => {
+    if (!seoAnalysis || !seoRecommendationsApplied || !setSuggestionsRef.current) return;
+    try {
+      if (suggestionsJson !== prevSuggestionsRef.current) {
+        setSuggestionsRef.current(suggestionsPayload);
+        debug.log('[BlogWriter] Forced Copilot suggestions sync after SEO recommendations applied', { count: suggestionsPayload.length });
+        prevSuggestionsRef.current = suggestionsJson;
+      }
+    } catch (e) {
+      console.error('Failed to push Copilot suggestions after SEO apply:', e);
+    }
+  }, [seoAnalysis, seoRecommendationsApplied, suggestionsJson, suggestionsPayload]);
+
+  return {
+    suggestions,
+    setSuggestionsRef,
+  };
+};
+
diff --git a/frontend/src/components/BlogWriter/BlogWriterUtils/useModalVisibility.ts b/frontend/src/components/BlogWriter/BlogWriterUtils/useModalVisibility.ts
new file mode 100644
index 00000000..e39f5313
--- /dev/null
+++ b/frontend/src/components/BlogWriter/BlogWriterUtils/useModalVisibility.ts
@@ -0,0 +1,61 @@
+import { useState, useEffect } from 'react';
+
+interface UseModalVisibilityProps {
+  mediumPolling: { isPolling: boolean };
+  rewritePolling: { isPolling: boolean };
+  outlinePolling: { isPolling: boolean };
+}
+
+export const useModalVisibility = ({
+  mediumPolling,
+  rewritePolling,
+  outlinePolling,
+}: UseModalVisibilityProps) => {
+  const [showModal, setShowModal] = useState(false);
+  const [modalStartTime, setModalStartTime] = useState<number | null>(null);
+  const [isMediumGenerationStarting, setIsMediumGenerationStarting] = useState(false);
+  const [showOutlineModal, setShowOutlineModal] = useState(false);
+
+  // Add minimum display time for modal
+  useEffect(() => {
+    if ((mediumPolling.isPolling || rewritePolling.isPolling || isMediumGenerationStarting) && !showModal) {
+      setShowModal(true);
+      setModalStartTime(Date.now());
+    } else if (!mediumPolling.isPolling && !rewritePolling.isPolling && !isMediumGenerationStarting && showModal) {
+      const elapsed = Date.now() - (modalStartTime || 0);
+      const minDisplayTime = 2000; // 2 seconds minimum
+      
+      if (elapsed < minDisplayTime) {
+        setTimeout(() => {
+          setShowModal(false);
+          setModalStartTime(null);
+        }, minDisplayTime - elapsed);
+      } else {
+        setShowModal(false);
+        setModalStartTime(null);
+      }
+    }
+  }, [mediumPolling.isPolling, rewritePolling.isPolling, isMediumGenerationStarting, showModal, modalStartTime]);
+
+  // Handle outline modal visibility
+  useEffect(() => {
+    if (outlinePolling.isPolling && !showOutlineModal) {
+      setShowOutlineModal(true);
+    } else if (!outlinePolling.isPolling && showOutlineModal) {
+      // Add a small delay to ensure user sees completion message
+      setTimeout(() => {
+        setShowOutlineModal(false);
+      }, 1000);
+    }
+  }, [outlinePolling.isPolling, showOutlineModal]);
+
+  return {
+    showModal,
+    setShowModal,
+    showOutlineModal,
+    setShowOutlineModal,
+    isMediumGenerationStarting,
+    setIsMediumGenerationStarting,
+  };
+};
+
diff --git a/frontend/src/components/BlogWriter/BlogWriterUtils/usePhaseActionHandlers.ts b/frontend/src/components/BlogWriter/BlogWriterUtils/usePhaseActionHandlers.ts
new file mode 100644
index 00000000..0f27c475
--- /dev/null
+++ b/frontend/src/components/BlogWriter/BlogWriterUtils/usePhaseActionHandlers.ts
@@ -0,0 +1,182 @@
+import { useCallback } from 'react';
+import { debug } from '../../../utils/debug';
+import { mediumBlogApi } from '../../../services/blogWriterApi';
+import { researchCache } from '../../../services/researchCache';
+import { blogWriterCache } from '../../../services/blogWriterCache';
+
+interface UsePhaseActionHandlersProps {
+  research: any;
+  outline: any[];
+  selectedTitle: string | null;
+  contentConfirmed: boolean;
+  sections: Record<string, string>;
+  navigateToPhase: (phase: string) => void;
+  handleOutlineConfirmed: () => void;
+  setIsMediumGenerationStarting: (starting: boolean) => void;
+  mediumPolling: any;
+  outlineGenRef: React.RefObject<any>;
+  setOutline: (outline: any[]) => void;
+  setContentConfirmed: (confirmed: boolean) => void;
+  setIsSEOMetadataModalOpen: (open: boolean) => void;
+  runSEOAnalysisDirect: () => string;
+  onOutlineComplete?: (outline: any) => void;
+  onContentComplete?: (sections: Record<string, string>) => void;
+}
+
+export const usePhaseActionHandlers = ({
+  research,
+  outline,
+  selectedTitle,
+  contentConfirmed,
+  sections,
+  navigateToPhase,
+  handleOutlineConfirmed,
+  setIsMediumGenerationStarting,
+  mediumPolling,
+  outlineGenRef,
+  setOutline,
+  setContentConfirmed,
+  setIsSEOMetadataModalOpen,
+  runSEOAnalysisDirect,
+  onOutlineComplete,
+  onContentComplete,
+}: UsePhaseActionHandlersProps) => {
+  const handleResearchAction = useCallback(() => {
+    navigateToPhase('research');
+    debug.log('[BlogWriter] Research action triggered - navigating to research phase');
+    // Note: Research caching is handled by ManualResearchForm component
+  }, [navigateToPhase]);
+
+  const handleOutlineAction = useCallback(async () => {
+    if (!research) {
+      alert('Please complete research first before generating an outline.');
+      return;
+    }
+    
+    // Check cache first (shared utility)
+    const researchKeywords = research.original_keywords || research.keyword_analysis?.primary || [];
+    const cachedOutline = blogWriterCache.getCachedOutline(researchKeywords);
+    
+    if (cachedOutline) {
+      debug.log('[BlogWriter] Using cached outline from localStorage', { sections: cachedOutline.outline.length });
+      setOutline(cachedOutline.outline);
+      if (onOutlineComplete) {
+        onOutlineComplete({ outline: cachedOutline.outline, title_options: cachedOutline.title_options });
+      }
+      navigateToPhase('outline');
+      return;
+    }
+    
+    navigateToPhase('outline');
+    if (outlineGenRef.current) {
+      try {
+        const result = await outlineGenRef.current.generateNow();
+        if (!result.success) {
+          alert(result.message || 'Failed to generate outline');
+        }
+      } catch (error) {
+        console.error('Outline generation failed:', error);
+        alert(`Outline generation failed: ${error instanceof Error ? error.message : 'Unknown error'}`);
+      }
+    }
+    debug.log('[BlogWriter] Outline action triggered');
+  }, [research, navigateToPhase, outlineGenRef, setOutline, onOutlineComplete]);
+
+  const handleContentAction = useCallback(async () => {
+    if (!outline || outline.length === 0) {
+      alert('Please generate and confirm an outline first.');
+      return;
+    }
+    if (!research) {
+      alert('Research data is required for content generation.');
+      return;
+    }
+    navigateToPhase('content');
+    
+    // Confirm outline first
+    handleOutlineConfirmed();
+    
+    // Check cache first (shared utility)
+    const outlineIds = outline.map(s => String(s.id));
+    const cachedContent = blogWriterCache.getCachedContent(outlineIds);
+    
+    if (cachedContent) {
+      debug.log('[BlogWriter] Using cached content from localStorage', { sections: Object.keys(cachedContent).length });
+      if (onContentComplete) {
+        onContentComplete(cachedContent);
+      }
+      return;
+    }
+    
+    // Also check if sections already exist in current state (shared utility)
+    if (blogWriterCache.contentExistsInState(sections || {}, outlineIds)) {
+      debug.log('[BlogWriter] Content already exists in state, skipping generation', { sections: Object.keys(sections || {}).length });
+      return;
+    }
+    
+    // If short/medium blog (<=1000 words), trigger content generation automatically
+    const target = Number(
+      research?.keyword_analysis?.blog_length || 
+      (research as any)?.word_count_target || 
+      localStorage.getItem('blog_length_target') || 
+      0
+    );
+    
+    if (target && target <= 1000) {
+      try {
+        setIsMediumGenerationStarting(true);
+        const payload = {
+          title: selectedTitle || (typeof window !== 'undefined' ? localStorage.getItem('blog_selected_title') : '') || outline[0]?.heading || 'Untitled',
+          sections: outline.map(s => ({
+            id: s.id,
+            heading: s.heading,
+            keyPoints: s.key_points,
+            subheadings: s.subheadings,
+            keywords: s.keywords,
+            targetWords: s.target_words,
+            references: s.references,
+          })),
+          globalTargetWords: target,
+          researchKeywords: research.original_keywords || research.keyword_analysis?.primary || [],
+        };
+
+        const { task_id } = await mediumBlogApi.startMediumGeneration(payload as any);
+        setIsMediumGenerationStarting(false);
+        mediumPolling.startPolling(task_id);
+        debug.log('[BlogWriter] Content action triggered - medium generation started', { task_id });
+      } catch (error) {
+        console.error('Content generation failed:', error);
+        setIsMediumGenerationStarting(false);
+        alert(`Content generation failed: ${error instanceof Error ? error.message : 'Unknown error'}`);
+      }
+    } else {
+      // For longer blogs, just confirm outline - user will use manual button
+      debug.log('[BlogWriter] Content action triggered - outline confirmed (manual content generation required)');
+    }
+  }, [outline, research, selectedTitle, sections, navigateToPhase, handleOutlineConfirmed, setIsMediumGenerationStarting, mediumPolling, onContentComplete]);
+
+  const handleSEOAction = useCallback(() => {
+    if (!contentConfirmed) {
+      // Mark content as confirmed when SEO action is clicked
+      setContentConfirmed(true);
+    }
+    navigateToPhase('seo');
+    runSEOAnalysisDirect();
+    debug.log('[BlogWriter] SEO action triggered');
+  }, [contentConfirmed, setContentConfirmed, navigateToPhase, runSEOAnalysisDirect]);
+
+  const handlePublishAction = useCallback(() => {
+    navigateToPhase('publish');
+    setIsSEOMetadataModalOpen(true);
+    debug.log('[BlogWriter] Publish action triggered - opening SEO metadata modal');
+  }, [navigateToPhase, setIsSEOMetadataModalOpen]);
+
+  return {
+    handleResearchAction,
+    handleOutlineAction,
+    handleContentAction,
+    handleSEOAction,
+    handlePublishAction,
+  };
+};
+
diff --git a/frontend/src/components/BlogWriter/BlogWriterUtils/usePhaseRestoration.ts b/frontend/src/components/BlogWriter/BlogWriterUtils/usePhaseRestoration.ts
new file mode 100644
index 00000000..06673b7c
--- /dev/null
+++ b/frontend/src/components/BlogWriter/BlogWriterUtils/usePhaseRestoration.ts
@@ -0,0 +1,67 @@
+import { useEffect } from 'react';
+import { debug } from '../../../utils/debug';
+
+interface UsePhaseRestorationProps {
+  copilotKitAvailable: boolean;
+  research: any;
+  phases: any[];
+  currentPhase: string;
+  navigateToPhase: (phase: string) => void;
+  setCurrentPhase: (phase: string) => void;
+}
+
+export const usePhaseRestoration = ({
+  copilotKitAvailable,
+  research,
+  phases,
+  currentPhase,
+  navigateToPhase,
+  setCurrentPhase,
+}: UsePhaseRestorationProps) => {
+  // When CopilotKit is unavailable and there's no research, ensure we're on research phase
+  useEffect(() => {
+    if (!copilotKitAvailable && !research && phases.length > 0 && currentPhase !== 'research') {
+      navigateToPhase('research');
+      debug.log('[BlogWriter] Auto-navigating to research phase (CopilotKit unavailable)');
+    }
+  }, [copilotKitAvailable, research, phases.length, currentPhase, navigateToPhase]);
+
+  // Restore phase from navigation state on mount (after subscription renewal)
+  // Note: The PricingPage restores the phase to localStorage before redirecting
+  // This effect ensures the phase is applied when BlogWriter loads
+  useEffect(() => {
+    try {
+      // Wait for phases to be initialized
+      if (phases.length === 0) {
+        return;
+      }
+      
+      // Check if we just returned from pricing page (has restored phase in localStorage)
+      const restoredPhase = localStorage.getItem('blogwriter_current_phase');
+      const userSelectedPhase = localStorage.getItem('blogwriter_user_selected_phase') === 'true';
+      
+      // Only restore if:
+      // 1. A phase was saved (restoredPhase exists)
+      // 2. User had manually selected a phase (indicates they were actively working)
+      // 3. The phase is different from current (to avoid unnecessary updates)
+      if (restoredPhase && userSelectedPhase && restoredPhase !== currentPhase) {
+        const targetPhase = phases.find(p => p.id === restoredPhase);
+        if (targetPhase && !targetPhase.disabled) {
+          console.log('[BlogWriter] Restoring phase from navigation state:', restoredPhase);
+          setCurrentPhase(restoredPhase);
+          // Phase restoration complete - the usePhaseNavigation hook will handle persistence
+        } else {
+          console.log('[BlogWriter] Restored phase is disabled or not found, keeping current phase:', {
+            restoredPhase,
+            currentPhase,
+            targetPhaseExists: !!targetPhase,
+            targetPhaseDisabled: targetPhase?.disabled
+          });
+        }
+      }
+    } catch (error) {
+      console.error('[BlogWriter] Failed to restore phase from navigation state:', error);
+    }
+  }, [phases, currentPhase, setCurrentPhase]);
+};
+
diff --git a/frontend/src/components/BlogWriter/BlogWriterUtils/useSEOManager.ts b/frontend/src/components/BlogWriter/BlogWriterUtils/useSEOManager.ts
new file mode 100644
index 00000000..23ee2e93
--- /dev/null
+++ b/frontend/src/components/BlogWriter/BlogWriterUtils/useSEOManager.ts
@@ -0,0 +1,245 @@
+import { useState, useRef, useEffect, useCallback } from 'react';
+import { debug } from '../../../utils/debug';
+import { blogWriterApi, BlogSEOActionableRecommendation } from '../../../services/blogWriterApi';
+
+interface UseSEOManagerProps {
+  sections: Record<string, string>;
+  research: any;
+  outline: any[];
+  selectedTitle: string | null;
+  contentConfirmed: boolean;
+  seoAnalysis: any;
+  currentPhase: string;
+  navigateToPhase: (phase: string) => void;
+  setContentConfirmed: (confirmed: boolean) => void;
+  setSeoAnalysis: (analysis: any) => void;
+  setSeoMetadata: (metadata: any) => void;
+  setSections: (sections: Record<string, string>) => void;
+  setSelectedTitle: (title: string | null) => void;
+  setContinuityRefresh: (timestamp: number) => void;
+  setFlowAnalysisCompleted: (completed: boolean) => void;
+  setFlowAnalysisResults: (results: any) => void;
+}
+
+export const useSEOManager = ({
+  sections,
+  research,
+  outline,
+  selectedTitle,
+  contentConfirmed,
+  seoAnalysis,
+  currentPhase,
+  navigateToPhase,
+  setContentConfirmed,
+  setSeoAnalysis,
+  setSeoMetadata,
+  setSections,
+  setSelectedTitle,
+  setContinuityRefresh,
+  setFlowAnalysisCompleted,
+  setFlowAnalysisResults,
+}: UseSEOManagerProps) => {
+  const [isSEOAnalysisModalOpen, setIsSEOAnalysisModalOpen] = useState(false);
+  const [isSEOMetadataModalOpen, setIsSEOMetadataModalOpen] = useState(false);
+  const [seoRecommendationsApplied, setSeoRecommendationsApplied] = useState(false);
+  const lastSEOModalOpenRef = useRef<number>(0);
+
+  // Helper: run same checks as analyzeSEO and open modal
+  const runSEOAnalysisDirect = useCallback((): string => {
+    const hasSections = !!sections && Object.keys(sections).length > 0;
+    const hasResearch = !!research && !!(research as any).keyword_analysis;
+    if (!hasSections) return "No blog content available for SEO analysis. Please generate content first.";
+    if (!hasResearch) return "Research data is required for SEO analysis. Please run research first.";
+    // Prevent rapid re-opens
+    const now = Date.now();
+    if (isSEOAnalysisModalOpen && now - lastSEOModalOpenRef.current < 1000) {
+      return "SEO analysis is already open.";
+    }
+    
+    // Mark content phase as done when user clicks "Next: Run SEO Analysis"
+    if (!contentConfirmed) {
+      setContentConfirmed(true);
+      debug.log('[BlogWriter] Content phase marked as done (SEO analysis triggered)');
+    }
+    
+    setSeoRecommendationsApplied(false);
+    if (!isSEOAnalysisModalOpen) {
+      setIsSEOAnalysisModalOpen(true);
+      lastSEOModalOpenRef.current = now;
+      debug.log('[BlogWriter] SEO modal opened (direct)');
+    }
+    return "Running SEO analysis of your blog content. This will analyze content structure, keyword optimization, readability, and provide actionable recommendations.";
+  }, [sections, research, isSEOAnalysisModalOpen, contentConfirmed, setContentConfirmed]);
+
+  const handleApplySeoRecommendations = useCallback(async (
+    recommendations: BlogSEOActionableRecommendation[]
+  ) => {
+    if (!outline || outline.length === 0) {
+      throw new Error('An outline is required before applying recommendations.');
+    }
+
+    const sectionPayload = outline.map((section) => ({
+      id: section.id,
+      heading: section.heading,
+      content: sections[section.id] ?? '',
+    }));
+
+    const response = await blogWriterApi.applySeoRecommendations({
+      title: selectedTitle || outline[0]?.heading || 'Untitled Blog',
+      sections: sectionPayload,
+      outline,
+      research: (research as any) || {},
+      recommendations,
+    });
+
+    if (!response.success) {
+      throw new Error(response.error || 'Failed to apply recommendations.');
+    }
+
+    if (!response.sections || !Array.isArray(response.sections)) {
+      throw new Error('Recommendation response did not include updated sections.');
+    }
+
+    // Update sections - create new object reference to trigger React re-render
+    const newSections: Record<string, string> = {};
+    response.sections.forEach((section) => {
+      if (section.id && section.content) {
+        newSections[section.id] = section.content;
+      }
+    });
+    
+    // Validate we have sections before updating
+    if (Object.keys(newSections).length === 0) {
+      throw new Error('No valid sections received from SEO recommendations application.');
+    }
+    
+    // Validate sections have actual content
+    const sectionsWithContent = Object.values(newSections).filter(c => c && c.trim().length > 0);
+    if (sectionsWithContent.length === 0) {
+      throw new Error('SEO recommendations resulted in empty sections. Please try again.');
+    }
+    
+    // Log detailed section info for debugging
+    const sectionIds = Object.keys(newSections);
+    const sectionSizes = sectionIds.map(id => ({ id, length: newSections[id]?.length || 0 }));
+    debug.log('[BlogWriter] Applied SEO recommendations: sections updated', { 
+      sectionCount: sectionIds.length,
+      sectionsWithContent: sectionsWithContent.length,
+      sectionIds: sectionIds,
+      sectionSizes: sectionSizes,
+      totalContentLength: Object.values(newSections).reduce((sum, c) => sum + (c?.length || 0), 0)
+    });
+    
+    // Update sections state
+    setSections(newSections);
+    
+    // Force a delay to ensure React processes the state update before proceeding
+    // This gives React time to re-render with new sections before phase navigation checks
+    await new Promise(resolve => setTimeout(resolve, 200));
+    
+    setContinuityRefresh(Date.now());
+    setFlowAnalysisCompleted(false);
+    setFlowAnalysisResults(null);
+
+    if (response.title && response.title !== selectedTitle) {
+      setSelectedTitle(response.title);
+    }
+
+    if (response.applied) {
+      setSeoAnalysis((prev: any) => prev ? { ...prev, applied_recommendations: response.applied } : prev);
+      debug.log('[BlogWriter] SEO analysis state updated with applied recommendations');
+    }
+
+    // Mark recommendations as applied (this will trigger phase navigation check)
+    // But we'll stay in SEO phase to show updated content
+    setSeoRecommendationsApplied(true);
+    debug.log('[BlogWriter] seoRecommendationsApplied set to true');
+    
+    // Ensure we stay in SEO phase to show updated content
+    // Force navigation to SEO phase if we're not already there (safeguard)
+    if (currentPhase !== 'seo') {
+      navigateToPhase('seo');
+      debug.log('[BlogWriter] Forced navigation to SEO phase after applying recommendations');
+    } else {
+      debug.log('[BlogWriter] Already in SEO phase, staying to show updated content');
+    }
+  }, [outline, sections, selectedTitle, research, setSections, setSelectedTitle, setContinuityRefresh, setFlowAnalysisCompleted, setFlowAnalysisResults, setSeoAnalysis, currentPhase, navigateToPhase]);
+
+  // Handle SEO analysis completion
+  const handleSEOAnalysisComplete = useCallback((analysis: any) => {
+    setSeoAnalysis(analysis);
+    debug.log('[BlogWriter] SEO analysis completed', { hasAnalysis: !!analysis });
+  }, [setSeoAnalysis]);
+
+  // Handle SEO modal close - mark SEO phase as done if not already marked
+  const handleSEOModalClose = useCallback(() => {
+    // Mark SEO phase as done when modal closes (even without applying recommendations)
+    if (!seoAnalysis) {
+      // Set a minimal valid seoAnalysis object to mark phase as complete
+      setSeoAnalysis({
+        success: true,
+        overall_score: 0,
+        category_scores: {},
+        analysis_summary: {
+          overall_grade: 'N/A',
+          status: 'Skipped',
+          strongest_category: 'N/A',
+          weakest_category: 'N/A',
+          key_strengths: [],
+          key_weaknesses: [],
+          ai_summary: 'SEO analysis was skipped by user'
+        },
+        actionable_recommendations: [],
+        generated_at: new Date().toISOString()
+      });
+      debug.log('[BlogWriter] SEO phase marked as done (modal closed without analysis)');
+    }
+    setIsSEOAnalysisModalOpen(false);
+    debug.log('[BlogWriter] SEO modal closed');
+  }, [seoAnalysis, setSeoAnalysis]);
+
+  // Mark SEO phase as completed when recommendations are applied
+  useEffect(() => {
+    if (seoRecommendationsApplied && seoAnalysis) {
+      // SEO phase is considered complete when recommendations are applied
+      // But stay in SEO phase to show updated content
+      debug.log('[BlogWriter] SEO recommendations applied, SEO phase marked as complete');
+      
+      // Ensure we stay in SEO phase to show updated content (override auto-progression)
+      if (currentPhase !== 'seo' && Object.keys(sections).length > 0) {
+        navigateToPhase('seo');
+        debug.log('[BlogWriter] Forced stay in SEO phase to show updated content');
+      }
+    }
+  }, [seoRecommendationsApplied, seoAnalysis, currentPhase, sections, navigateToPhase]);
+
+  const confirmBlogContent = useCallback(() => {
+    debug.log('[BlogWriter] Blog content confirmed by user');
+    setContentConfirmed(true);
+    setSeoRecommendationsApplied(false);
+    navigateToPhase('seo');
+    setTimeout(() => {
+      setIsSEOAnalysisModalOpen(true);
+      debug.log('[BlogWriter] SEO modal opened (confirm→direct)');
+    }, 0);
+    return "✅ Blog content has been confirmed! Running SEO analysis now.";
+  }, [setContentConfirmed, navigateToPhase]);
+
+  return {
+    isSEOAnalysisModalOpen,
+    setIsSEOAnalysisModalOpen,
+    isSEOMetadataModalOpen,
+    setIsSEOMetadataModalOpen,
+    seoRecommendationsApplied,
+    setSeoRecommendationsApplied,
+    lastSEOModalOpenRef,
+    runSEOAnalysisDirect,
+    handleApplySeoRecommendations,
+    handleSEOAnalysisComplete,
+    handleSEOModalClose,
+    confirmBlogContent,
+  };
+};
+
+export type SEOManagerReturn = ReturnType<typeof useSEOManager>;
+
diff --git a/frontend/src/components/BlogWriter/ManualContentButton.tsx b/frontend/src/components/BlogWriter/ManualContentButton.tsx
new file mode 100644
index 00000000..2e4ba807
--- /dev/null
+++ b/frontend/src/components/BlogWriter/ManualContentButton.tsx
@@ -0,0 +1,113 @@
+import React, { useState } from 'react';
+import { Button, CircularProgress } from '@mui/material';
+import { mediumBlogApi } from '../../services/blogWriterApi';
+import { BlogOutlineSection, BlogResearchResponse } from '../../services/blogWriterApi';
+
+interface ManualContentButtonProps {
+  /**
+   * The confirmed outline sections
+   */
+  outline: BlogOutlineSection[];
+  /**
+   * The research data
+   */
+  research: BlogResearchResponse;
+  /**
+   * Blog title (optional)
+   */
+  blogTitle?: string;
+  /**
+   * Existing sections content (optional)
+   */
+  sections?: Record<string, string>;
+  /**
+   * Callback when content generation starts
+   */
+  onGenerationStart?: (taskId: string) => void;
+}
+
+/**
+ * Manual content generation button that works independently of CopilotKit
+ * Triggers medium blog generation via mediumBlogApi
+ */
+export const ManualContentButton: React.FC<ManualContentButtonProps> = ({
+  outline,
+  research,
+  blogTitle,
+  sections,
+  onGenerationStart,
+}) => {
+  const [isGenerating, setIsGenerating] = useState(false);
+  const [error, setError] = useState<string | null>(null);
+
+  const handleGenerate = async () => {
+    if (!outline || outline.length === 0) {
+      alert('Please confirm an outline first before generating content.');
+      return;
+    }
+
+    if (!research) {
+      alert('Research data is required for content generation.');
+      return;
+    }
+
+    setIsGenerating(true);
+    setError(null);
+
+    try {
+      const payload = {
+        outline,
+        research,
+        title: blogTitle || outline[0]?.heading || 'Blog Post',
+        existing_sections: sections || {},
+      };
+
+      const { task_id } = await mediumBlogApi.startMediumGeneration(payload as any);
+      
+      if (task_id) {
+        onGenerationStart?.(task_id);
+      } else {
+        throw new Error('Failed to start content generation - no task ID returned');
+      }
+    } catch (error) {
+      const errorMessage = error instanceof Error ? error.message : 'Unknown error';
+      setError(errorMessage);
+      alert(`Content generation failed: ${errorMessage}`);
+      setIsGenerating(false);
+    }
+  };
+
+  return (
+    <div style={{ padding: '20px', textAlign: 'center' }}>
+      <h3 style={{ margin: '0 0 16px 0', color: '#333' }}>Generate Blog Content</h3>
+      <p style={{ margin: '0 0 20px 0', color: '#666', fontSize: '14px' }}>
+        Generate full content for all sections in your confirmed outline.
+      </p>
+      
+      <Button
+        variant="contained"
+        color="primary"
+        size="large"
+        onClick={handleGenerate}
+        disabled={!outline || outline.length === 0 || !research || isGenerating}
+        startIcon={isGenerating ? <CircularProgress size={20} color="inherit" /> : null}
+        sx={{
+          minWidth: 200,
+          py: 1.5,
+          px: 4,
+        }}
+      >
+        {isGenerating ? 'Generating Content...' : '📝 Generate Content'}
+      </Button>
+
+      {error && (
+        <p style={{ margin: '12px 0 0 0', color: '#d32f2f', fontSize: '14px' }}>
+          {error}
+        </p>
+      )}
+    </div>
+  );
+};
+
+export default ManualContentButton;
+
diff --git a/frontend/src/components/BlogWriter/ManualOutlineButton.tsx b/frontend/src/components/BlogWriter/ManualOutlineButton.tsx
new file mode 100644
index 00000000..c8ee20e5
--- /dev/null
+++ b/frontend/src/components/BlogWriter/ManualOutlineButton.tsx
@@ -0,0 +1,111 @@
+import React, { useState } from 'react';
+import { Button, CircularProgress } from '@mui/material';
+
+interface ManualOutlineButtonProps {
+  /**
+   * Ref to OutlineGenerator component with generateNow() method
+   */
+  outlineGenRef: React.RefObject<{ 
+    generateNow: () => Promise<{ 
+      success: boolean; 
+      message?: string; 
+      task_id?: string;
+      cached?: boolean;
+      outline?: any[];
+      title_options?: string[];
+    }> 
+  }>;
+  /**
+   * Whether research is available (required for outline generation)
+   */
+  hasResearch: boolean;
+  /**
+   * Callback when outline generation starts
+   */
+  onGenerationStart?: (taskId: string) => void;
+}
+
+/**
+ * Manual outline generation button that works independently of CopilotKit
+ * Calls the generateNow() method from OutlineGenerator ref
+ */
+export const ManualOutlineButton: React.FC<ManualOutlineButtonProps> = ({
+  outlineGenRef,
+  hasResearch,
+  onGenerationStart,
+}) => {
+  const [isGenerating, setIsGenerating] = useState(false);
+  const [error, setError] = useState<string | null>(null);
+
+  const handleGenerate = async () => {
+    if (!hasResearch) {
+      alert('Please complete research first before generating an outline.');
+      return;
+    }
+
+    if (!outlineGenRef.current) {
+      alert('Outline generator is not available. Please refresh the page.');
+      return;
+    }
+
+    setIsGenerating(true);
+    setError(null);
+
+    try {
+      const result = await outlineGenRef.current.generateNow();
+      
+      if (result.success) {
+        if (result.cached && result.outline) {
+          // Handle cached result - outline is already available, no need to poll
+          console.log('[ManualOutlineButton] Cached outline used', { sections: result.outline.length });
+          // The outline should be set by the parent component handling the cache
+        } else if (result.task_id) {
+          onGenerationStart?.(result.task_id);
+        }
+      } else {
+        setError(result.message || 'Failed to generate outline');
+        alert(result.message || 'Failed to generate outline. Please try again.');
+      }
+    } catch (error) {
+      const errorMessage = error instanceof Error ? error.message : 'Unknown error';
+      setError(errorMessage);
+      alert(`Outline generation failed: ${errorMessage}`);
+    } finally {
+      setIsGenerating(false);
+    }
+  };
+
+  return (
+    <div style={{ padding: '20px', textAlign: 'center' }}>
+      <h3 style={{ margin: '0 0 16px 0', color: '#333' }}>Create Your Outline</h3>
+      <p style={{ margin: '0 0 20px 0', color: '#666', fontSize: '14px' }}>
+        Generate an AI-powered outline based on your research.
+      </p>
+      
+      <Button
+        variant="contained"
+        color="primary"
+        size="large"
+        onClick={handleGenerate}
+        disabled={!hasResearch || isGenerating}
+        startIcon={isGenerating ? <CircularProgress size={20} color="inherit" /> : null}
+        sx={{
+          minWidth: 200,
+          py: 1.5,
+          px: 4,
+        }}
+      >
+        {isGenerating ? 'Generating Outline...' : '🧩 Generate Outline'}
+      </Button>
+
+      {error && (
+        <p style={{ margin: '12px 0 0 0', color: '#d32f2f', fontSize: '14px' }}>
+          {error}
+        </p>
+      )}
+    </div>
+  );
+};
+
+export default ManualOutlineButton;
+
diff --git a/frontend/src/components/BlogWriter/ManualResearchForm.tsx b/frontend/src/components/BlogWriter/ManualResearchForm.tsx
new file mode 100644
index 00000000..fd891d0b
--- /dev/null
+++ b/frontend/src/components/BlogWriter/ManualResearchForm.tsx
@@ -0,0 +1,184 @@
+import React, { useState, useRef } from 'react';
+import { blogWriterApi, BlogResearchRequest, BlogResearchResponse } from '../../services/blogWriterApi';
+import { useResearchPolling } from '../../hooks/usePolling';
+import ResearchProgressModal from './ResearchProgressModal';
+import { researchCache } from '../../services/researchCache';
+
+interface ManualResearchFormProps {
+  onResearchComplete?: (research: BlogResearchResponse) => void;
+}
+
+/**
+ * Manual research form component that works independently of CopilotKit
+ * Extracted from ResearchAction.tsx for use when CopilotKit is unavailable
+ */
+export const ManualResearchForm: React.FC<ManualResearchFormProps> = ({ onResearchComplete }) => {
+  const [currentTaskId, setCurrentTaskId] = useState<string | null>(null);
+  const [currentMessage, setCurrentMessage] = useState<string>('');
+  const [showProgressModal, setShowProgressModal] = useState<boolean>(false);
+  const [isSubmitting, setIsSubmitting] = useState(false);
+  
+  // Refs for form inputs (uncontrolled, avoids typing issues)
+  const keywordsRef = useRef<HTMLInputElement | null>(null);
+  const blogLengthRef = useRef<HTMLSelectElement | null>(null);
+
+  const polling = useResearchPolling({
+    onProgress: (message) => {
+      setCurrentMessage(message);
+    },
+    onComplete: (result) => {
+      if (result && result.keywords) {
+        researchCache.cacheResult(
+          result.keywords,
+          result.industry || 'General',
+          result.target_audience || 'General',
+          result
+        );
+      }
+      
+      onResearchComplete?.(result);
+      setCurrentTaskId(null);
+      setCurrentMessage('');
+      setShowProgressModal(false);
+      setIsSubmitting(false);
+    },
+    onError: (error) => {
+      console.error('Research polling error:', error);
+      setCurrentTaskId(null);
+      setCurrentMessage('');
+      setShowProgressModal(false);
+      setIsSubmitting(false);
+    }
+  });
+
+  const handleSubmit = async () => {
+    const keywords = (keywordsRef.current?.value || '').trim();
+    const blogLength = blogLengthRef.current?.value || '1000';
+    
+    if (!keywords) {
+      alert('Please enter keywords or a topic for research.');
+      return;
+    }
+
+    setIsSubmitting(true);
+    
+    try {
+      const keywordList = keywords.includes(',') 
+        ? keywords.split(',').map(k => k.trim()).filter(Boolean) 
+        : [keywords];
+      
+      // Check cache first
+      const cachedResult = researchCache.getCachedResult(keywordList, 'General', 'General');
+      if (cachedResult) {
+        onResearchComplete?.(cachedResult);
+        setIsSubmitting(false);
+        return;
+      }
+      
+      const payload: BlogResearchRequest = { 
+        keywords: keywordList, 
+        industry: 'General', 
+        target_audience: 'General',
+        word_count_target: parseInt(blogLength)
+      };
+      
+      const { task_id } = await blogWriterApi.startResearch(payload);
+      setCurrentTaskId(task_id);
+      setShowProgressModal(true);
+      polling.startPolling(task_id);
+    } catch (error) {
+      console.error('Research failed:', error);
+      alert(`Research failed: ${error instanceof Error ? error.message : 'Unknown error'}`);
+      setIsSubmitting(false);
+    }
+  };
+
+  return (
+    <>
+      <div style={{ padding: '20px', backgroundColor: '#f8f9fa', borderRadius: '12px', border: '1px solid #e0e0e0', margin: '8px 0' }}>
+        <h4 style={{ margin: '0 0 16px 0', color: '#333' }}>🔍 Let's Research Your Blog Topic</h4>
+        <p style={{ margin: '0 0 16px 0', color: '#666', fontSize: '14px' }}>
+          What keywords and information would you like to use for your research? Please also specify the desired length of the blog post.
+        </p>
+        
+        <div style={{ marginBottom: '16px' }}>
+          <label style={{ display: 'block', marginBottom: '8px', fontWeight: '500', color: '#333' }}>Keywords or Topic *</label>
+          <input
+            type="text"
+            id="research-keywords-input"
+            placeholder="e.g., artificial intelligence, machine learning, AI trends"
+            ref={keywordsRef}
+            disabled={isSubmitting}
+            style={{ 
+              width: '100%', 
+              padding: '12px', 
+              border: '1px solid #ddd', 
+              borderRadius: '6px', 
+              fontSize: '14px', 
+              boxSizing: 'border-box',
+              opacity: isSubmitting ? 0.6 : 1
+            }}
+          />
+        </div>
+        
+        <div style={{ marginBottom: '16px' }}>
+          <label style={{ display: 'block', marginBottom: '8px', fontWeight: '500', color: '#333' }}>Blog Length (words)</label>
+          <select
+            id="research-blog-length-select"
+            defaultValue="1000"
+            ref={blogLengthRef}
+            disabled={isSubmitting}
+            style={{ 
+              width: '100%', 
+              padding: '12px', 
+              border: '1px solid #ddd', 
+              borderRadius: '6px', 
+              fontSize: '14px', 
+              boxSizing: 'border-box',
+              opacity: isSubmitting ? 0.6 : 1
+            }}
+          >
+            <option value="500">500 words (Short blog)</option>
+            <option value="1000">1000 words (Medium blog)</option>
+            <option value="1500">1500 words (Long blog)</option>
+            <option value="2000">2000 words (Comprehensive blog)</option>
+          </select>
+        </div>
+        
+        <div style={{ display: 'flex', gap: '12px', justifyContent: 'flex-end' }}>
+          <button
+            onClick={handleSubmit}
+            disabled={isSubmitting}
+            style={{ 
+              padding: '12px 24px', 
+              backgroundColor: isSubmitting ? '#ccc' : '#1976d2', 
+              color: 'white', 
+              border: 'none', 
+              borderRadius: '6px', 
+              fontSize: '14px', 
+              fontWeight: '500', 
+              cursor: isSubmitting ? 'not-allowed' : 'pointer',
+              opacity: isSubmitting ? 0.7 : 1
+            }}
+          >
+            {isSubmitting ? '⏳ Starting Research...' : '🚀 Start Research'}
+          </button>
+        </div>
+      </div>
+
+      {showProgressModal && (
+        <ResearchProgressModal
+          open={showProgressModal}
+          title="Research in progress"
+          status={polling.currentStatus}
+          messages={polling.progressMessages}
+          error={polling.error}
+          onClose={() => setShowProgressModal(false)}
+        />
+      )}
+    </>
+  );
+};
+
+export default ManualResearchForm;
+
diff --git a/frontend/src/components/BlogWriter/OutlineFeedbackForm.tsx b/frontend/src/components/BlogWriter/OutlineFeedbackForm.tsx
index 25114c53..c83fa1e5 100644
--- a/frontend/src/components/BlogWriter/OutlineFeedbackForm.tsx
+++ b/frontend/src/components/BlogWriter/OutlineFeedbackForm.tsx
@@ -363,6 +363,21 @@ export const OutlineFeedbackForm: React.FC<OutlineFeedbackFormProps> = ({
         );
         
         if (target && target <= 1000) {
+          // Check cache first (shared utility)
+          const { blogWriterCache } = await import('../../services/blogWriterCache');
+          const outlineIds = outline.map(s => String(s.id));
+          const cachedContent = blogWriterCache.getCachedContent(outlineIds);
+          
+          if (cachedContent) {
+            console.log('[OutlineFeedbackForm] Using cached content', { sections: Object.keys(cachedContent).length });
+            // Content is already cached, skip API call
+            return {
+              success: true,
+              message: 'Content is already available from cache.',
+              cached: true
+            };
+          }
+          
           // Show modal immediately when medium generation is triggered
           onMediumGenerationTriggered?.();
           // Build payload for medium generation
@@ -386,13 +401,61 @@ export const OutlineFeedbackForm: React.FC<OutlineFeedbackFormProps> = ({
           // Notify parent to start polling for the medium generation task
           onMediumGenerationStarted?.(task_id);
 
-          // Return message so the copilot shows feedback; UI will display modal via BlogWriter
-          return {
-            success: true,
-            message: `✅ Outline confirmed. Medium generation started (Task: ${task_id}). You can monitor progress in the modal.`,
-            task_id,
-            action_taken: 'outline_confirmed_medium_generation_started'
-          };
+          // Poll once immediately to check for immediate failures (e.g., subscription errors)
+          try {
+            const initialStatus = await mediumBlogApi.pollMediumGeneration(task_id);
+            
+            // Check if task already failed with subscription error
+            if (initialStatus.status === 'failed' && (initialStatus.error_status === 429 || initialStatus.error_status === 402)) {
+              const errorData = initialStatus.error_data || {};
+              const errorMessage = errorData.message || errorData.error || initialStatus.error || 'Subscription limit exceeded';
+              
+              // Return error to CopilotKit so it shows in chat
+              return {
+                success: false,
+                message: `❌ Medium generation failed: ${errorMessage}`,
+                error: errorMessage,
+                error_type: 'subscription_limit',
+                provider: errorData.provider || 'unknown',
+                suggestion: 'Please renew your subscription to continue generating content.',
+                action_taken: 'outline_confirmed_medium_generation_failed'
+              };
+            }
+            
+            // Task started successfully, continue polling in background
+            return {
+              success: true,
+              message: `✅ Outline confirmed. Medium generation started (Task: ${task_id}). You can monitor progress in the modal.`,
+              task_id,
+              action_taken: 'outline_confirmed_medium_generation_started'
+            };
+          } catch (pollError: any) {
+            // Check if polling error is a subscription error (HTTP 429/402)
+            if (pollError?.response?.status === 429 || pollError?.response?.status === 402) {
+              const errorData = pollError.response?.data || {};
+              const errorMessage = errorData.message || errorData.error || 'Subscription limit exceeded';
+              
+              return {
+                success: false,
+                message: `❌ Medium generation failed: ${errorMessage}`,
+                error: errorMessage,
+                error_type: 'subscription_limit',
+                provider: errorData.provider || 'unknown',
+                suggestion: 'Please renew your subscription to continue generating content.',
+                action_taken: 'outline_confirmed_medium_generation_failed'
+              };
+            }
+            
+            // Other polling errors - still return success since task was started
+            // The polling will handle the error in the background
+            console.warn('Initial poll check failed, but task was started:', pollError);
+            return {
+              success: true,
+              message: `✅ Outline confirmed. Medium generation started (Task: ${task_id}). You can monitor progress in the modal.`,
+              task_id,
+              action_taken: 'outline_confirmed_medium_generation_started'
+            };
+          }
         }
 
         return {
diff --git a/frontend/src/components/BlogWriter/OutlineGenerator.tsx b/frontend/src/components/BlogWriter/OutlineGenerator.tsx
index 5de43603..917642d1 100644
--- a/frontend/src/components/BlogWriter/OutlineGenerator.tsx
+++ b/frontend/src/components/BlogWriter/OutlineGenerator.tsx
@@ -1,6 +1,7 @@
 import React, { forwardRef, useImperativeHandle } from 'react';
 import { useCopilotAction } from '@copilotkit/react-core';
 import { blogWriterApi, BlogResearchResponse } from '../../services/blogWriterApi';
+import { blogWriterCache } from '../../services/blogWriterCache';
 
 interface OutlineGeneratorProps {
   research: BlogResearchResponse | null;
@@ -23,6 +24,22 @@ export const OutlineGenerator = forwardRef<any, OutlineGeneratorProps>(({
       if (!research) {
         return { success: false, message: 'No research yet. Please research a topic first.' };
       }
+      
+      // Check cache first (shared utility)
+      const researchKeywords = research.original_keywords || research.keyword_analysis?.primary || [];
+      const cachedOutline = blogWriterCache.getCachedOutline(researchKeywords);
+      
+      if (cachedOutline) {
+        console.log('[OutlineGenerator] Using cached outline', { sections: cachedOutline.outline.length });
+        // Return cached result - caller should handle setting outline state
+        return { 
+          success: true, 
+          cached: true,
+          outline: cachedOutline.outline,
+          title_options: cachedOutline.title_options
+        };
+      }
+      
       try {
         onModalShow?.();
         const { task_id } = await blogWriterApi.startOutlineGeneration({ research });
@@ -44,6 +61,21 @@ export const OutlineGenerator = forwardRef<any, OutlineGeneratorProps>(({
         return { success: false, message: 'No research yet. Please research a topic first.' };
       }
       
+      // Check cache first (shared utility)
+      const researchKeywords = research.original_keywords || research.keyword_analysis?.primary || [];
+      const cachedOutline = blogWriterCache.getCachedOutline(researchKeywords);
+      
+      if (cachedOutline) {
+        console.log('[OutlineGenerator] Using cached outline from CopilotKit action', { sections: cachedOutline.outline.length });
+        return {
+          success: true,
+          message: `✅ Outline already exists! ${cachedOutline.outline.length} sections loaded from cache.`,
+          cached: true,
+          outline: cachedOutline.outline,
+          title_options: cachedOutline.title_options
+        };
+      }
+      
       try {
         // Show progress modal immediately when user clicks "Create outline"
         onModalShow?.();
diff --git a/frontend/src/components/BlogWriter/PhaseNavigation.tsx b/frontend/src/components/BlogWriter/PhaseNavigation.tsx
index 0ea7c412..04a3a9ed 100644
--- a/frontend/src/components/BlogWriter/PhaseNavigation.tsx
+++ b/frontend/src/components/BlogWriter/PhaseNavigation.tsx
@@ -10,17 +10,80 @@ export interface Phase {
   disabled: boolean;
 }
 
+export interface PhaseActionHandlers {
+  onResearchAction?: () => void;      // Show research form
+  onOutlineAction?: () => void;       // Generate outline
+  onContentAction?: () => void;       // Confirm outline + generate content
+  onSEOAction?: () => void;           // Run SEO analysis
+  onPublishAction?: () => void;       // Generate SEO metadata or publish
+}
+
 interface PhaseNavigationProps {
   phases: Phase[];
   onPhaseClick: (phaseId: string) => void;
   currentPhase: string;
+  copilotKitAvailable?: boolean;
+  actionHandlers?: PhaseActionHandlers;
+  // State for determining which actions to show
+  hasResearch?: boolean;
+  hasOutline?: boolean;
+  outlineConfirmed?: boolean;
+  hasContent?: boolean;
+  contentConfirmed?: boolean;
+  hasSEOAnalysis?: boolean;
+  hasSEOMetadata?: boolean;
 }
 
 export const PhaseNavigation: React.FC<PhaseNavigationProps> = ({
   phases,
   onPhaseClick,
-  currentPhase
+  currentPhase,
+  copilotKitAvailable = true,
+  actionHandlers,
+  hasResearch = false,
+  hasOutline = false,
+  outlineConfirmed = false,
+  hasContent = false,
+  contentConfirmed = false,
+  hasSEOAnalysis = false,
+  hasSEOMetadata = false,
 }) => {
+  // Determine which action to show for each phase when CopilotKit is unavailable
+  const getActionForPhase = (phaseId: string): { label: string; handler: (() => void) | null } => {
+    if (copilotKitAvailable || !actionHandlers) {
+      return { label: '', handler: null };
+    }
+
+    switch (phaseId) {
+      case 'research':
+        if (!hasResearch) {
+          return { label: 'Start Research', handler: actionHandlers.onResearchAction || null };
+        }
+        break;
+      case 'outline':
+        if (hasResearch && !hasOutline) {
+          return { label: 'Create Outline', handler: actionHandlers.onOutlineAction || null };
+        }
+        break;
+      case 'content':
+        if (hasOutline && !outlineConfirmed) {
+          return { label: 'Confirm & Generate Content', handler: actionHandlers.onContentAction || null };
+        }
+        break;
+      case 'seo':
+        if (hasContent && contentConfirmed && !hasSEOAnalysis) {
+          return { label: 'Run SEO Analysis', handler: actionHandlers.onSEOAction || null };
+        }
+        break;
+      case 'publish':
+        if (hasSEOAnalysis && !hasSEOMetadata) {
+          return { label: 'Generate SEO Metadata', handler: actionHandlers.onPublishAction || null };
+        }
+        break;
+    }
+    return { label: '', handler: null };
+  };
+
   return (
     <div style={{
       display: 'flex',
@@ -33,53 +96,103 @@ export const PhaseNavigation: React.FC<PhaseNavigationProps> = ({
         const isCurrent = phase.current;
         const isCompleted = phase.completed;
         const isDisabled = phase.disabled;
+        const action = getActionForPhase(phase.id);
+        // Show action button when:
+        // 1. CopilotKit is unavailable
+        // 2. Action handler exists
+        // 3. Phase is not disabled
+        // 4. Show for current phase OR next actionable phase (not completed)
+        //    For research phase specifically, always show if no research exists
+        const isResearchPhase = phase.id === 'research' && !hasResearch;
+        const showAction = !copilotKitAvailable && action.handler && !isDisabled && (
+          isCurrent || 
+          (!isCompleted && !isDisabled) ||
+          isResearchPhase
+        );
         
         return (
-          <button
-            key={phase.id}
-            onClick={() => !isDisabled && onPhaseClick(phase.id)}
-            disabled={isDisabled}
-            style={{
-              display: 'flex',
-              alignItems: 'center',
-              gap: '6px',
-              padding: '8px 12px',
-              borderRadius: '20px',
-              border: 'none',
-              fontSize: '14px',
-              fontWeight: '500',
-              cursor: isDisabled ? 'not-allowed' : 'pointer',
-              transition: 'all 0.2s ease',
-              backgroundColor: isCurrent 
-                ? '#1976d2' 
-                : isCompleted 
-                  ? '#4caf50' 
-                  : isDisabled 
-                    ? '#f5f5f5' 
-                    : '#e3f2fd',
-              color: isCurrent 
-                ? 'white' 
-                : isCompleted 
+          <div key={phase.id} style={{ display: 'flex', alignItems: 'center', gap: '6px' }}>
+            <button
+              onClick={() => !isDisabled && onPhaseClick(phase.id)}
+              disabled={isDisabled}
+              style={{
+                display: 'flex',
+                alignItems: 'center',
+                gap: '6px',
+                padding: '8px 12px',
+                borderRadius: '20px',
+                border: 'none',
+                fontSize: '14px',
+                fontWeight: '500',
+                cursor: isDisabled ? 'not-allowed' : 'pointer',
+                transition: 'all 0.2s ease',
+                backgroundColor: isCurrent 
+                  ? '#1976d2' 
+                  : isCompleted 
+                    ? '#4caf50' 
+                    : isDisabled 
+                      ? '#f5f5f5' 
+                      : '#e3f2fd',
+                color: isCurrent 
                   ? 'white' 
-                  : isDisabled 
-                    ? '#999' 
-                    : '#1976d2',
-              opacity: isDisabled ? 0.6 : 1,
-              boxShadow: isCurrent ? '0 2px 4px rgba(25, 118, 210, 0.3)' : 'none',
-              transform: isCurrent ? 'translateY(-1px)' : 'none'
-            }}
-            title={phase.disabled ? `Complete ${phase.name} first` : phase.description}
-          >
-            <span style={{ fontSize: '16px' }}>
-              {phase.icon}
-            </span>
-            <span>{phase.name}</span>
-            {isCompleted && !isCurrent && (
-              <span style={{ fontSize: '12px', marginLeft: '4px' }}>
-                ✓
+                  : isCompleted 
+                    ? 'white' 
+                    : isDisabled 
+                      ? '#999' 
+                      : '#1976d2',
+                opacity: isDisabled ? 0.6 : 1,
+                boxShadow: isCurrent ? '0 2px 4px rgba(25, 118, 210, 0.3)' : 'none',
+                transform: isCurrent ? 'translateY(-1px)' : 'none'
+              }}
+              title={phase.disabled ? `Complete ${phase.name} first` : phase.description}
+            >
+              <span style={{ fontSize: '16px' }}>
+                {phase.icon}
               </span>
+              <span>{phase.name}</span>
+              {isCompleted && !isCurrent && (
+                <span style={{ fontSize: '12px', marginLeft: '4px' }}>
+                  ✓
+                </span>
+              )}
+            </button>
+            
+            {showAction && (
+              <button
+                onClick={(e) => {
+                  e.stopPropagation();
+                  action.handler?.();
+                }}
+                style={{
+                  display: 'flex',
+                  alignItems: 'center',
+                  gap: '4px',
+                  padding: '6px 12px',
+                  borderRadius: '16px',
+                  border: '1px solid #1976d2',
+                  fontSize: '12px',
+                  fontWeight: '600',
+                  cursor: 'pointer',
+                  backgroundColor: '#1976d2',
+                  color: 'white',
+                  transition: 'all 0.2s ease',
+                  boxShadow: '0 2px 4px rgba(25, 118, 210, 0.2)',
+                }}
+                onMouseEnter={(e) => {
+                  e.currentTarget.style.backgroundColor = '#1565c0';
+                  e.currentTarget.style.transform = 'translateY(-1px)';
+                }}
+                onMouseLeave={(e) => {
+                  e.currentTarget.style.backgroundColor = '#1976d2';
+                  e.currentTarget.style.transform = 'none';
+                }}
+                title={`${action.label} (Chat unavailable - click to proceed)`}
+              >
+                <span style={{ fontSize: '12px' }}>▶</span>
+                <span>{action.label}</span>
+              </button>
             )}
-          </button>
+          </div>
         );
       })}
     </div>
diff --git a/frontend/src/components/BlogWriter/Publisher.tsx b/frontend/src/components/BlogWriter/Publisher.tsx
index f54a541b..88ef43ba 100644
--- a/frontend/src/components/BlogWriter/Publisher.tsx
+++ b/frontend/src/components/BlogWriter/Publisher.tsx
@@ -105,19 +105,27 @@ export const Publisher: React.FC<PublisherProps> = ({
 
     try {
       // Publish using same endpoint as WixTestPage
-      // Note: Wix requires category/tag IDs (UUIDs), not names
-      // For now, skip categories/tags until we implement ID lookup/creation
+      // Backend will lookup/create category and tag IDs from names if needed
       const response = await apiClient.post('/api/wix/test/publish/real', {
         title: title,
         content: md, // Use markdown, backend converts it
         cover_image_url: coverImageUrl,
-        // TODO: Lookup/create category IDs from metadata?.blog_categories
-        // TODO: Lookup/create tag IDs from metadata?.blog_tags
-        category_ids: undefined,
-        tag_ids: undefined,
+        // Pass category/tag names - backend will lookup existing or create new ones
+        category_names: metadata?.blog_categories || [],
+        tag_names: metadata?.blog_tags || [],
         publish: true,
         access_token: accessToken,
-        member_id: undefined // Let backend derive from token
+        member_id: undefined, // Let backend derive from token
+        seo_metadata: metadata ? {
+          seo_title: metadata.seo_title,
+          meta_description: metadata.meta_description,
+          focus_keyword: metadata.focus_keyword,
+          blog_tags: metadata.blog_tags || [], // Used for SEO keywords
+          social_hashtags: metadata.social_hashtags || [],
+          open_graph: metadata.open_graph || {},
+          twitter_card: metadata.twitter_card || {},
+          canonical_url: metadata.canonical_url
+        } : undefined
       });
 
       if (response.data.success) {
diff --git a/frontend/src/components/BlogWriter/ResearchComponents/ResearchSources.tsx b/frontend/src/components/BlogWriter/ResearchComponents/ResearchSources.tsx
index 41628458..4a053cde 100644
--- a/frontend/src/components/BlogWriter/ResearchComponents/ResearchSources.tsx
+++ b/frontend/src/components/BlogWriter/ResearchComponents/ResearchSources.tsx
@@ -1,4 +1,4 @@
-import React, { useEffect, useState } from 'react';
+import React, { useState } from 'react';
 import { BlogResearchResponse } from '../../../services/blogWriterApi';
 
 interface ResearchSourcesProps {
@@ -187,24 +187,6 @@ const KeywordChipGroup: React.FC<KeywordChipGroupProps> = ({
 };
 
 export const ResearchSources: React.FC<ResearchSourcesProps> = ({ research }) => {
-  const [showWebSearchHelp, setShowWebSearchHelp] = useState(false);
-
-  // Fix search widget overflow after render
-  useEffect(() => {
-    if (research.search_widget) {
-      const searchWidget = document.querySelector('[data-search-widget]');
-      if (searchWidget) {
-        const allElements = searchWidget.querySelectorAll('*');
-        allElements.forEach((el: any) => {
-          el.style.maxWidth = '100%';
-          el.style.overflow = 'hidden';
-          el.style.wordWrap = 'break-word';
-          el.style.whiteSpace = 'normal';
-          el.style.boxSizing = 'border-box';
-        });
-      }
-    }
-  }, [research.search_widget]);
 
   const renderCredibilityScore = (score: number | undefined) => {
     const safeScore = score ?? 0.8; // Default to 0.8 if undefined
@@ -454,135 +436,17 @@ export const ResearchSources: React.FC<ResearchSourcesProps> = ({ research }) =>
         </div>
       )}
         
-        {/* Interactive Web Search - Moved from Header */}
+        {/* Google Search Suggestions - Per Google Display Requirements */}
         {research.search_widget && (
-          <div style={{ marginBottom: '20px', width: '100%', overflow: 'hidden' }}>
-            <div style={{ display: 'flex', alignItems: 'center', gap: '8px', marginBottom: '12px', position: 'relative' }}>
-              <h4 style={{ margin: 0, color: '#555', fontSize: '16px' }}>
-                🔍 Explore More Research Topics
-              </h4>
-              {/* Help Icon for Web Search */}
-              <span 
-                onClick={() => setShowWebSearchHelp(!showWebSearchHelp)}
-                style={{
-                  fontSize: '14px',
-                  color: '#9ca3af',
-                  cursor: 'pointer',
-                  padding: '4px',
-                  borderRadius: '50%',
-                  transition: 'all 0.2s ease',
-                  display: 'flex',
-                  alignItems: 'center',
-                  justifyContent: 'center',
-                  minWidth: '24px',
-                  minHeight: '24px'
-                }}
-                onMouseEnter={(e) => {
-                  e.currentTarget.style.color = '#6b7280';
-                  e.currentTarget.style.backgroundColor = '#f3f4f6';
-                }}
-                onMouseLeave={(e) => {
-                  e.currentTarget.style.color = '#9ca3af';
-                  e.currentTarget.style.backgroundColor = 'transparent';
-                }}
-              >
-                ❓
-              </span>
-              
-              {/* Help Tooltip for Web Search */}
-              {showWebSearchHelp && (
-                <div style={{
-                  position: 'absolute',
-                  top: '100%',
-                  left: '0',
-                  marginTop: '8px',
-                  backgroundColor: '#1f2937',
-                  color: '#f9fafb',
-                  padding: '12px 16px',
-                  borderRadius: '8px',
-                  fontSize: '12px',
-                  lineHeight: '1.5',
-                  maxWidth: '300px',
-                  boxShadow: '0 10px 15px -3px rgba(0, 0, 0, 0.1), 0 4px 6px -2px rgba(0, 0, 0, 0.05)',
-                  zIndex: 1000,
-                  border: '1px solid #374151'
-                }}>
-                  <div style={{ fontWeight: '600', marginBottom: '4px', color: '#f3f4f6' }}>
-                    Research Enhancement
-                  </div>
-                  <div style={{ color: '#d1d5db' }}>
-                    Click on any search suggestion below to explore additional research topics and gather more insights for your blog. These searches will open in a new tab to help you discover trending topics, expert opinions, and current statistics.
-                  </div>
-                  {/* Tooltip arrow */}
-                  <div style={{
-                    position: 'absolute',
-                    bottom: '100%',
-                    left: '20px',
-                    width: 0,
-                    height: 0,
-                    borderLeft: '6px solid transparent',
-                    borderRight: '6px solid transparent',
-                    borderBottom: '6px solid #1f2937'
-                  }} />
-                </div>
-              )}
-            </div>
-            <div style={{
-              border: '1px solid #e0e0e0',
-              borderRadius: '8px',
-              padding: '16px',
-              backgroundColor: '#fafafa',
-              maxHeight: '400px',
-              overflow: 'auto',
-              width: '100%',
-              maxWidth: '100%',
-              boxSizing: 'border-box',
-              overflowX: 'hidden',
-              position: 'relative'
-            }}
-            onClick={(e) => {
-              // Make all links open in new tabs
-              const target = e.target as HTMLElement;
-              if (target.tagName === 'A' || target.closest('a')) {
-                const link = target.tagName === 'A' ? target as HTMLAnchorElement : target.closest('a') as HTMLAnchorElement;
-                if (link && link.href) {
-                  link.target = '_blank';
-                  link.rel = 'noopener noreferrer';
-                }
-              }
-            }}>
-              <div 
-                data-search-widget
-                dangerouslySetInnerHTML={{ __html: research.search_widget }}
-                style={{ 
-                  fontSize: '14px',
-                  width: '100%',
-                  maxWidth: '100%',
-                  overflow: 'hidden',
-                  overflowX: 'hidden',
-                  wordWrap: 'break-word',
-                  overflowWrap: 'break-word',
-                  whiteSpace: 'normal',
-                  display: 'block',
-                  position: 'relative'
-                }}
-              />
-              
-              {/* Custom CSS to make Google icon larger */}
-              <style>
-                {`
-                  [data-search-widget] svg {
-                    width: 24px !important;
-                    height: 24px !important;
-                  }
-                  [data-search-widget] .logo-light,
-                  [data-search-widget] .logo-dark {
-                    width: 24px !important;
-                    height: 24px !important;
-                  }
-                `}
-              </style>
-            </div>
+          <div style={{ 
+            marginBottom: '24px', 
+            width: '100%',
+            position: 'relative'
+          }}>
+            {/* Google Search Widget - Display exactly as provided without modifications */}
+            <div 
+              dangerouslySetInnerHTML={{ __html: research.search_widget }}
+            />
           </div>
         )}
         
diff --git a/frontend/src/components/BlogWriter/SEOAnalysisModal.tsx b/frontend/src/components/BlogWriter/SEOAnalysisModal.tsx
index f6cf64c4..46ab2d70 100644
--- a/frontend/src/components/BlogWriter/SEOAnalysisModal.tsx
+++ b/frontend/src/components/BlogWriter/SEOAnalysisModal.tsx
@@ -27,7 +27,7 @@ import {
   Avatar,
   CircularProgress
 } from '@mui/material';
-import { apiClient } from '../../api/client';
+import { apiClient, triggerSubscriptionError } from '../../api/client';
 import { 
   CheckCircle, 
   Cancel, 
@@ -308,7 +308,28 @@ export const SEOAnalysisModal: React.FC<SEOAnalysisModalProps> = ({
         onAnalysisComplete(convertedResult);
       }
 
-    } catch (err) {
+    } catch (err: any) {
+      console.error('SEO analysis failed:', err);
+      
+      // Check if this is a subscription error (429/402) and trigger global subscription modal
+      const status = err?.response?.status;
+      if (status === 429 || status === 402) {
+        console.log('SEOAnalysisModal: Detected subscription error, triggering global handler', {
+          status,
+          data: err?.response?.data
+        });
+        const handled = triggerSubscriptionError(err);
+        if (handled) {
+          console.log('SEOAnalysisModal: Global subscription error handler triggered successfully');
+          // Don't set local error - let the global modal handle it
+          setIsAnalyzing(false);
+          return;
+        } else {
+          console.warn('SEOAnalysisModal: Global subscription error handler did not handle the error');
+        }
+      }
+      
+      // For non-subscription errors, show local error message
       setError(err instanceof Error ? err.message : 'Analysis failed');
       setIsAnalyzing(false);
     }
diff --git a/frontend/src/components/BlogWriter/SEOMetadataModal.tsx b/frontend/src/components/BlogWriter/SEOMetadataModal.tsx
index 028d89ad..c7d2da50 100644
--- a/frontend/src/components/BlogWriter/SEOMetadataModal.tsx
+++ b/frontend/src/components/BlogWriter/SEOMetadataModal.tsx
@@ -36,7 +36,7 @@ import {
   Tag as TagIcon,
   Refresh as RefreshIcon
 } from '@mui/icons-material';
-import { apiClient } from '../../api/client';
+import { apiClient, triggerSubscriptionError } from '../../api/client';
 
 // Import metadata display components
 import { CoreMetadataTab } from './SEO/MetadataDisplay/CoreMetadataTab';
@@ -219,8 +219,28 @@ export const SEOMetadataModal: React.FC<SEOMetadataModalProps> = ({
       setEditableMetadata(result);
       console.log('📊 Metadata result set:', result);
 
-    } catch (err) {
+    } catch (err: any) {
       console.error('❌ SEO metadata generation failed:', err);
+      
+      // Check if this is a subscription error (429/402) and trigger global subscription modal
+      const status = err?.response?.status;
+      if (status === 429 || status === 402) {
+        console.log('SEOMetadataModal: Detected subscription error, triggering global handler', {
+          status,
+          data: err?.response?.data
+        });
+        const handled = triggerSubscriptionError(err);
+        if (handled) {
+          console.log('SEOMetadataModal: Global subscription error handler triggered successfully');
+          // Don't set local error - let the global modal handle it
+          setIsGenerating(false);
+          return;
+        } else {
+          console.warn('SEOMetadataModal: Global subscription error handler did not handle the error');
+        }
+      }
+      
+      // For non-subscription errors, show local error message
       setError(err instanceof Error ? err.message : 'Failed to generate SEO metadata');
     } finally {
       setIsGenerating(false);
diff --git a/frontend/src/components/OnboardingWizard/common/usePlatformConnections.ts b/frontend/src/components/OnboardingWizard/common/usePlatformConnections.ts
index 1013c3db..340f3f1e 100644
--- a/frontend/src/components/OnboardingWizard/common/usePlatformConnections.ts
+++ b/frontend/src/components/OnboardingWizard/common/usePlatformConnections.ts
@@ -53,6 +53,21 @@ export const usePlatformConnections = () => {
 
   const handleWixConnect = async () => {
     try {
+      // Store current page URL BEFORE redirecting (critical for proper redirect back)
+      // This ensures we can redirect back to the correct page (e.g., Blog Writer) after OAuth
+      const currentUrl = window.location.href;
+      try {
+        // Only store if not already set (allows WixConnectModal to override if needed)
+        if (!sessionStorage.getItem('wix_oauth_redirect')) {
+          sessionStorage.setItem('wix_oauth_redirect', currentUrl);
+          console.log('[Wix OAuth] Stored redirect URL:', currentUrl);
+        } else {
+          console.log('[Wix OAuth] Redirect URL already set, keeping existing:', sessionStorage.getItem('wix_oauth_redirect'));
+        }
+      } catch (e) {
+        console.warn('[Wix OAuth] Failed to store redirect URL:', e);
+      }
+
       // Use the working Wix OAuth flow from WixTestPage
       const wixClient = createClient({
         auth: OAuthStrategy({ clientId: '75d88e36-1c76-4009-b769-15f4654556df' })
diff --git a/frontend/src/components/Pricing/PricingPage.tsx b/frontend/src/components/Pricing/PricingPage.tsx
index 895157c3..ab10cdd3 100644
--- a/frontend/src/components/Pricing/PricingPage.tsx
+++ b/frontend/src/components/Pricing/PricingPage.tsx
@@ -51,6 +51,7 @@ import {
 } from '@mui/icons-material';
 import { useNavigate } from 'react-router-dom';
 import { apiClient } from '../../api/client';
+import { restoreNavigationState, saveCurrentPhaseForTool } from '../../utils/navigationState';
 
 interface SubscriptionPlan {
   id: number;
@@ -114,8 +115,15 @@ const PricingPage: React.FC = () => {
   };
 
   const handleSubscribe = async (planId: number) => {
+    console.log('[PricingPage] handleSubscribe called', { planId });
+    
     const plan = plans.find(p => p.id === planId);
-    if (!plan) return;
+    if (!plan) {
+      console.error('[PricingPage] ❌ Plan not found for ID:', planId);
+      return;
+    }
+
+    console.log('[PricingPage] Selected plan:', { id: plan.id, name: plan.name, tier: plan.tier });
 
     // Get user_id from localStorage (set by Clerk auth)
     const userId = localStorage.getItem('user_id');
@@ -123,18 +131,20 @@ const PricingPage: React.FC = () => {
     // Check if user is signed in
     if (!userId || userId === 'anonymous' || userId === '') {
       // User not signed in, show sign-in prompt
-      console.warn('PricingPage: User not signed in, showing prompt');
+      console.warn('[PricingPage] User not signed in, showing prompt');
       setShowSignInPrompt(true);
       return;
     }
 
     // For alpha testing, only allow Free and Basic plans (Pro features not ready)
     if (plan.tier !== 'free' && plan.tier !== 'basic') {
+      console.error('[PricingPage] Plan tier not available:', plan.tier);
       setError('This plan is not available for alpha testing');
       return;
     }
 
     if (plan.tier === 'free') {
+      console.log('[PricingPage] Processing Free plan subscription directly');
       // For free plan, just create subscription
       try {
         setSubscribing(true);
@@ -164,23 +174,38 @@ const PricingPage: React.FC = () => {
       }
     } else {
       // For Basic plan, show payment modal
+      console.log('[PricingPage] Opening payment modal for Basic plan', { planId, planName: plan.name });
+      setSelectedPlan(planId);  // ✅ Set selected plan before opening modal
       setPaymentModalOpen(true);
     }
   };
 
   const handlePaymentConfirm = async () => {
-    if (!selectedPlan) return;
+    console.log('[PricingPage] handlePaymentConfirm called', { selectedPlan, yearlyBilling });
+    
+    if (!selectedPlan) {
+      console.error('[PricingPage] ❌ No selectedPlan set - cannot proceed with subscription');
+      setError('No plan selected. Please select a plan and try again.');
+      return;
+    }
 
     try {
       setSubscribing(true);
       const userId = localStorage.getItem('user_id') || 'anonymous';
 
+      console.log('[PricingPage] Making subscription API call:', {
+        url: `/api/subscription/subscribe/${userId}`,
+        plan_id: selectedPlan,
+        billing_cycle: yearlyBilling ? 'yearly' : 'monthly',
+        userId
+      });
+
       const response = await apiClient.post(`/api/subscription/subscribe/${userId}`, {
         plan_id: selectedPlan,
         billing_cycle: yearlyBilling ? 'yearly' : 'monthly'
       });
 
-      console.log('Subscription renewed successfully:', response.data);
+      console.log('[PricingPage] ✅ Subscription renewed successfully:', response.data);
 
       // Refresh subscription status immediately
       window.dispatchEvent(new CustomEvent('subscription-updated'));
@@ -223,13 +248,26 @@ const PricingPage: React.FC = () => {
         // If not complete, redirect to onboarding; otherwise to dashboard
         const onboardingComplete = localStorage.getItem('onboarding_complete') === 'true';
         if (onboardingComplete) {
-          // Try to go back to where the user was (e.g., blog writer)
-          // If no history, go to dashboard
-          const referrer = sessionStorage.getItem('subscription_referrer');
-          if (referrer && referrer !== '/pricing') {
-            navigate(referrer);
+          // Restore navigation state (path, phase, tool) if available
+          const navState = restoreNavigationState();
+          
+          if (navState && navState.path && navState.path !== '/pricing') {
+            // Restore phase if applicable (e.g., Blog Writer)
+            if (navState.tool === 'blog-writer' && navState.phase) {
+              saveCurrentPhaseForTool('blog-writer', navState.phase);
+              console.log('[PricingPage] Restored Blog Writer phase:', navState.phase);
+            }
+            
+            console.log('[PricingPage] Redirecting to saved navigation state:', navState);
+            navigate(navState.path);
           } else {
-            navigate('/dashboard');
+            // Fallback: try legacy referrer
+            const referrer = sessionStorage.getItem('subscription_referrer');
+            if (referrer && referrer !== '/pricing') {
+              navigate(referrer);
+            } else {
+              navigate('/dashboard');
+            }
           }
         } else {
           navigate('/onboarding');
diff --git a/frontend/src/components/Research/ResearchWizard.tsx b/frontend/src/components/Research/ResearchWizard.tsx
new file mode 100644
index 00000000..a37ffd8a
--- /dev/null
+++ b/frontend/src/components/Research/ResearchWizard.tsx
@@ -0,0 +1,216 @@
+import React, { useEffect } from 'react';
+import { useResearchWizard } from './hooks/useResearchWizard';
+import { useResearchExecution } from './hooks/useResearchExecution';
+import { StepKeyword } from './steps/StepKeyword';
+import { StepOptions } from './steps/StepOptions';
+import { StepProgress } from './steps/StepProgress';
+import { StepResults } from './steps/StepResults';
+import { ResearchWizardProps } from './types/research.types';
+
+export const ResearchWizard: React.FC<ResearchWizardProps> = ({ 
+  onComplete,
+  onCancel,
+  initialKeywords,
+  initialIndustry,
+}) => {
+  const wizard = useResearchWizard(initialKeywords, initialIndustry);
+  const execution = useResearchExecution();
+
+  // Handle results from execution
+  useEffect(() => {
+    if (execution.result && !execution.isExecuting) {
+      wizard.updateState({ results: execution.result });
+      if (wizard.state.currentStep === 3) {
+        wizard.nextStep();
+      }
+    }
+  }, [execution.result, execution.isExecuting]);
+
+  // Handle completion callback
+  useEffect(() => {
+    if (wizard.state.results && onComplete) {
+      onComplete(wizard.state.results);
+    }
+  }, [wizard.state.results, onComplete]);
+
+  const renderStep = () => {
+    const stepProps = {
+      state: wizard.state,
+      onUpdate: wizard.updateState,
+      onNext: wizard.nextStep,
+      onBack: wizard.prevStep,
+    };
+
+    switch (wizard.state.currentStep) {
+      case 1:
+        return <StepKeyword {...stepProps} />;
+      case 2:
+        return <StepOptions {...stepProps} />;
+      case 3:
+        return <StepProgress {...stepProps} />;
+      case 4:
+        return <StepResults {...stepProps} />;
+      default:
+        return <StepKeyword {...stepProps} />;
+    }
+  };
+
+  return (
+    <div style={{ 
+      minHeight: '100vh', 
+      backgroundColor: '#f5f5f5',
+      padding: '20px',
+    }}>
+      {/* Wizard Container */}
+      <div style={{
+        maxWidth: '1200px',
+        margin: '0 auto',
+        backgroundColor: 'white',
+        borderRadius: '12px',
+        boxShadow: '0 2px 8px rgba(0,0,0,0.1)',
+        overflow: 'hidden',
+      }}>
+        {/* Header */}
+        <div style={{
+          backgroundColor: '#1976d2',
+          color: 'white',
+          padding: '24px',
+          borderBottom: '1px solid #e0e0e0',
+        }}>
+          <div style={{ display: 'flex', justifyContent: 'space-between', alignItems: 'center' }}>
+            <div>
+              <h1 style={{ margin: 0, fontSize: '24px' }}>Research Wizard</h1>
+              <p style={{ margin: '8px 0 0 0', fontSize: '14px', opacity: 0.9 }}>
+                Step {wizard.state.currentStep} of {wizard.maxSteps}
+              </p>
+            </div>
+            {onCancel && (
+              <button
+                onClick={onCancel}
+                style={{
+                  padding: '8px 16px',
+                  backgroundColor: 'rgba(255,255,255,0.2)',
+                  color: 'white',
+                  border: '1px solid rgba(255,255,255,0.3)',
+                  borderRadius: '6px',
+                  cursor: 'pointer',
+                  fontSize: '14px',
+                }}
+              >
+                Cancel
+              </button>
+            )}
+          </div>
+        </div>
+
+        {/* Progress Bar */}
+        <div style={{
+          backgroundColor: '#f0f0f0',
+          height: '4px',
+          position: 'relative',
+        }}>
+          <div
+            style={{
+              backgroundColor: '#1976d2',
+              height: '100%',
+              width: `${(wizard.state.currentStep / wizard.maxSteps) * 100}%`,
+              transition: 'width 0.3s ease',
+            }}
+          />
+        </div>
+
+        {/* Step Indicators */}
+        <div style={{
+          display: 'flex',
+          justifyContent: 'space-around',
+          padding: '20px 40px',
+          borderBottom: '1px solid #e0e0e0',
+        }}>
+          {[1, 2, 3, 4].map(step => (
+            <div key={step} style={{ display: 'flex', flexDirection: 'column', alignItems: 'center', position: 'relative' }}>
+              <div style={{
+                width: '40px',
+                height: '40px',
+                borderRadius: '50%',
+                backgroundColor: step <= wizard.state.currentStep ? '#1976d2' : '#e0e0e0',
+                color: step <= wizard.state.currentStep ? 'white' : '#999',
+                display: 'flex',
+                alignItems: 'center',
+                justifyContent: 'center',
+                fontWeight: 'bold',
+                fontSize: '16px',
+                marginBottom: '8px',
+                transition: 'all 0.3s ease',
+              }}>
+                {step < wizard.state.currentStep ? '✓' : step}
+              </div>
+              <span style={{
+                fontSize: '12px',
+                color: step <= wizard.state.currentStep ? '#1976d2' : '#999',
+                fontWeight: step === wizard.state.currentStep ? '600' : 'normal',
+              }}>
+                {step === 1 && 'Setup'}
+                {step === 2 && 'Options'}
+                {step === 3 && 'Research'}
+                {step === 4 && 'Results'}
+              </span>
+            </div>
+          ))}
+        </div>
+
+        {/* Content */}
+        <div style={{ padding: '24px' }}>
+          {renderStep()}
+        </div>
+
+        {/* Navigation Footer */}
+        {wizard.state.currentStep <= 2 && (
+          <div style={{
+            padding: '20px 24px',
+            borderTop: '1px solid #e0e0e0',
+            display: 'flex',
+            justifyContent: 'space-between',
+            alignItems: 'center',
+            backgroundColor: '#fafafa',
+          }}>
+            <button
+              onClick={wizard.prevStep}
+              disabled={wizard.isFirstStep}
+              style={{
+                padding: '10px 20px',
+                backgroundColor: wizard.isFirstStep ? '#f0f0f0' : 'white',
+                color: wizard.isFirstStep ? '#999' : '#333',
+                border: wizard.isFirstStep ? '1px solid #e0e0e0' : '1px solid #ddd',
+                borderRadius: '6px',
+                cursor: wizard.isFirstStep ? 'not-allowed' : 'pointer',
+                fontSize: '14px',
+              }}
+            >
+              ← Back
+            </button>
+
+            <button
+              onClick={wizard.nextStep}
+              disabled={!wizard.canGoNext()}
+              style={{
+                padding: '10px 24px',
+                backgroundColor: wizard.canGoNext() ? '#1976d2' : '#e0e0e0',
+                color: wizard.canGoNext() ? 'white' : '#999',
+                border: 'none',
+                borderRadius: '6px',
+                cursor: wizard.canGoNext() ? 'pointer' : 'not-allowed',
+                fontSize: '14px',
+                fontWeight: '600',
+              }}
+            >
+              {wizard.isLastStep ? 'Finish' : 'Next →'}
+            </button>
+          </div>
+        )}
+      </div>
+    </div>
+  );
+};
+
+export default ResearchWizard;
+
diff --git a/frontend/src/components/Research/hooks/useResearchExecution.ts b/frontend/src/components/Research/hooks/useResearchExecution.ts
new file mode 100644
index 00000000..70fd048e
--- /dev/null
+++ b/frontend/src/components/Research/hooks/useResearchExecution.ts
@@ -0,0 +1,82 @@
+import { useState, useCallback } from 'react';
+import { blogWriterApi, BlogResearchRequest, BlogResearchResponse } from '../../../services/blogWriterApi';
+import { useResearchPolling } from '../../../hooks/usePolling';
+import { researchCache } from '../../../services/researchCache';
+import { WizardState } from '../types/research.types';
+
+export const useResearchExecution = () => {
+  const [isExecuting, setIsExecuting] = useState(false);
+  const [error, setError] = useState<string | null>(null);
+
+  const polling = useResearchPolling({
+    onComplete: (result) => {
+      if (result && result.keywords) {
+        researchCache.cacheResult(
+          result.keywords,
+          'General',
+          'General',
+          result
+        );
+      }
+      setIsExecuting(false);
+    },
+    onError: (error) => {
+      console.error('Research polling error:', error);
+      setError(error);
+      setIsExecuting(false);
+    }
+  });
+
+  const executeResearch = useCallback(async (state: WizardState): Promise<string | null> => {
+    setIsExecuting(true);
+    setError(null);
+
+    try {
+      // Check cache first
+      const cachedResult = researchCache.getCachedResult(
+        state.keywords,
+        state.industry,
+        state.targetAudience
+      );
+      
+      if (cachedResult) {
+        setIsExecuting(false);
+        return 'cached';
+      }
+
+      const payload: BlogResearchRequest = {
+        keywords: state.keywords,
+        industry: state.industry,
+        target_audience: state.targetAudience,
+        research_mode: state.researchMode,
+        config: state.config,
+      };
+
+      const { task_id } = await blogWriterApi.startResearch(payload);
+      polling.startPolling(task_id);
+      return task_id;
+    } catch (err) {
+      const errorMessage = err instanceof Error ? err.message : 'Unknown error';
+      setError(errorMessage);
+      setIsExecuting(false);
+      return null;
+    }
+  }, [polling]);
+
+  const stopExecution = useCallback(() => {
+    polling.stopPolling();
+    setIsExecuting(false);
+    setError(null);
+  }, [polling]);
+
+  return {
+    executeResearch,
+    stopExecution,
+    isExecuting,
+    error,
+    progressMessages: polling.progressMessages,
+    currentStatus: polling.currentStatus,
+    result: polling.result,
+  };
+};
+
diff --git a/frontend/src/components/Research/hooks/useResearchWizard.ts b/frontend/src/components/Research/hooks/useResearchWizard.ts
new file mode 100644
index 00000000..144e193e
--- /dev/null
+++ b/frontend/src/components/Research/hooks/useResearchWizard.ts
@@ -0,0 +1,116 @@
+import { useState, useCallback, useEffect } from 'react';
+import { WizardState, WizardStepProps } from '../types/research.types';
+import { ResearchMode, ResearchConfig, BlogResearchResponse } from '../../../services/blogWriterApi';
+
+const WIZARD_STATE_KEY = 'alwrity_research_wizard_state';
+const MAX_STEPS = 4;
+
+const defaultState: WizardState = {
+  currentStep: 1,
+  keywords: [],
+  industry: 'General',
+  targetAudience: 'General',
+  researchMode: 'basic' as ResearchMode,
+  config: {
+    mode: 'basic',
+    provider: 'google',
+    max_sources: 10,
+    include_statistics: true,
+    include_expert_quotes: true,
+    include_competitors: true,
+    include_trends: true,
+  },
+  results: null,
+};
+
+export const useResearchWizard = (initialKeywords?: string[], initialIndustry?: string) => {
+  const [state, setState] = useState<WizardState>(() => {
+    // Try to load from localStorage first
+    const saved = localStorage.getItem(WIZARD_STATE_KEY);
+    if (saved) {
+      try {
+        const parsed = JSON.parse(saved);
+        return parsed;
+      } catch (e) {
+        console.warn('Failed to parse saved wizard state', e);
+      }
+    }
+    
+    // Use defaults or initial values
+    return {
+      ...defaultState,
+      keywords: initialKeywords || [],
+      industry: initialIndustry || defaultState.industry,
+    };
+  });
+
+  // Persist state to localStorage
+  useEffect(() => {
+    if (state.currentStep > 1) {
+      localStorage.setItem(WIZARD_STATE_KEY, JSON.stringify(state));
+    }
+  }, [state]);
+
+  const updateState = useCallback((updates: Partial<WizardState>) => {
+    setState(prev => ({ ...prev, ...updates }));
+  }, []);
+
+  const nextStep = useCallback(() => {
+    setState(prev => {
+      if (prev.currentStep >= MAX_STEPS) return prev;
+      return { ...prev, currentStep: prev.currentStep + 1 };
+    });
+  }, []);
+
+  const prevStep = useCallback(() => {
+    setState(prev => {
+      if (prev.currentStep <= 1) return prev;
+      return { ...prev, currentStep: prev.currentStep - 1 };
+    });
+  }, []);
+
+  const reset = useCallback(() => {
+    const resetState = {
+      ...defaultState,
+      keywords: initialKeywords || [],
+      industry: initialIndustry || defaultState.industry,
+    };
+    setState(resetState);
+    localStorage.removeItem(WIZARD_STATE_KEY);
+  }, [initialKeywords, initialIndustry]);
+
+  const clearResults = useCallback(() => {
+    setState(prev => ({ ...prev, results: null }));
+  }, []);
+
+  const canGoNext = useCallback((): boolean => {
+    switch (state.currentStep) {
+      case 1:
+        return state.keywords.length > 0 && state.keywords.every(k => k.trim().length > 0);
+      case 2:
+        return true; // Mode selection always allowed
+      case 3:
+        return false; // Progress can't be skipped
+      case 4:
+        return false; // Results can't be skipped
+      default:
+        return false;
+    }
+  }, [state]);
+
+  return {
+    state,
+    updateState,
+    nextStep,
+    prevStep,
+    reset,
+    clearResults,
+    canGoNext,
+    isFirstStep: state.currentStep === 1,
+    isLastStep: state.currentStep === MAX_STEPS,
+    maxSteps: MAX_STEPS,
+  };
+};
+
+export type UseResearchWizardReturn = ReturnType<typeof useResearchWizard>;
+
diff --git a/frontend/src/components/Research/index.tsx b/frontend/src/components/Research/index.tsx
new file mode 100644
index 00000000..dc75eb3e
--- /dev/null
+++ b/frontend/src/components/Research/index.tsx
@@ -0,0 +1,5 @@
+export { ResearchWizard } from './ResearchWizard';
+export { useResearchWizard } from './hooks/useResearchWizard';
+export { useResearchExecution } from './hooks/useResearchExecution';
+export * from './types/research.types';
+
diff --git a/frontend/src/components/Research/integrations/BlogWriterAdapter.tsx b/frontend/src/components/Research/integrations/BlogWriterAdapter.tsx
new file mode 100644
index 00000000..e509011e
--- /dev/null
+++ b/frontend/src/components/Research/integrations/BlogWriterAdapter.tsx
@@ -0,0 +1,72 @@
+/**
+ * Blog Writer Integration Adapter for Research Component
+ * 
+ * This adapter provides a simple way to integrate the ResearchWizard
+ * into the BlogWriter's research phase.
+ */
+
+import React from 'react';
+import { ResearchWizard } from '../ResearchWizard';
+import { BlogResearchResponse } from '../../../services/blogWriterApi';
+
+interface BlogWriterResearchAdapterProps {
+  onResearchComplete: (research: BlogResearchResponse) => void;
+  onCancel?: () => void;
+  initialKeywords?: string[];
+  initialIndustry?: string;
+}
+
+/**
+ * Adapter component that wraps ResearchWizard for BlogWriter integration.
+ * Provides a clean interface for switching between CopilotKit and wizard-based research.
+ */
+export const BlogWriterResearchAdapter: React.FC<BlogWriterResearchAdapterProps> = ({
+  onResearchComplete,
+  onCancel,
+  initialKeywords,
+  initialIndustry,
+}) => {
+  return (
+    <div style={{ 
+      height: '100%', 
+      display: 'flex', 
+      flexDirection: 'column',
+      backgroundColor: 'white',
+    }}>
+      <ResearchWizard
+        onComplete={onResearchComplete}
+        onCancel={onCancel}
+        initialKeywords={initialKeywords}
+        initialIndustry={initialIndustry}
+      />
+    </div>
+  );
+};
+
+export default BlogWriterResearchAdapter;
+
+/**
+ * USAGE EXAMPLE:
+ * 
+ * In BlogWriter.tsx, replace the research phase content with:
+ * 
+ * {currentPhase === 'research' && !research && (
+ *   <BlogWriterResearchAdapter
+ *     onResearchComplete={(res) => {
+ *       handleResearchComplete(res);
+ *       // Optionally auto-advance to outline phase
+ *       navigateToPhase('outline');
+ *     }}
+ *     onCancel={() => {
+ *       // Navigate back to dashboard
+ *       navigateToPhase('research');
+ *     }}
+ *     initialKeywords={[]}
+ *     initialIndustry="General"
+ *   />
+ * )}
+ * 
+ * Note: This maintains backward compatibility. The existing CopilotKit/manual
+ * research flow continues to work. This provides an alternative UI option.
+ */
+
diff --git a/frontend/src/components/Research/steps/StepKeyword.tsx b/frontend/src/components/Research/steps/StepKeyword.tsx
new file mode 100644
index 00000000..9e512c1c
--- /dev/null
+++ b/frontend/src/components/Research/steps/StepKeyword.tsx
@@ -0,0 +1,133 @@
+import React, { useEffect } from 'react';
+import { WizardStepProps } from '../types/research.types';
+
+const industries = [
+  'General',
+  'Technology',
+  'Business',
+  'Marketing',
+  'Finance',
+  'Healthcare',
+  'Education',
+  'Real Estate',
+  'Entertainment',
+  'Food & Beverage',
+  'Travel',
+  'Fashion',
+  'Sports',
+  'Science',
+  'Law',
+  'Other',
+];
+
+export const StepKeyword: React.FC<WizardStepProps> = ({ state, onUpdate }) => {
+  const handleKeywordsChange = (e: React.ChangeEvent<HTMLTextAreaElement>) => {
+    const value = e.target.value;
+    const keywords = value.split(',').map(k => k.trim()).filter(Boolean);
+    onUpdate({ keywords });
+  };
+
+  const handleIndustryChange = (e: React.ChangeEvent<HTMLSelectElement>) => {
+    onUpdate({ industry: e.target.value });
+  };
+
+  const handleAudienceChange = (e: React.ChangeEvent<HTMLInputElement>) => {
+    onUpdate({ targetAudience: e.target.value });
+  };
+
+  const keywordText = state.keywords.join(', ');
+
+  return (
+    <div style={{ padding: '24px', maxWidth: '800px', margin: '0 auto' }}>
+      <h2 style={{ marginBottom: '8px', color: '#333' }}>🔍 Research Setup</h2>
+      <p style={{ marginBottom: '24px', color: '#666', fontSize: '15px' }}>
+        Enter your keywords, industry, and target audience to start research.
+      </p>
+
+      {/* Keywords Input */}
+      <div style={{ marginBottom: '20px' }}>
+        <label style={{ display: 'block', marginBottom: '8px', fontWeight: '600', color: '#555' }}>
+          Keywords *
+        </label>
+        <textarea
+          value={keywordText}
+          onChange={handleKeywordsChange}
+          placeholder="e.g., AI in marketing, automation tools, customer engagement"
+          rows={4}
+          style={{
+            width: '100%',
+            padding: '12px',
+            border: '1px solid #ddd',
+            borderRadius: '8px',
+            fontSize: '14px',
+            fontFamily: 'inherit',
+            resize: 'vertical',
+            boxSizing: 'border-box',
+          }}
+        />
+        <p style={{ marginTop: '4px', fontSize: '12px', color: '#888' }}>
+          Separate multiple keywords with commas
+        </p>
+      </div>
+
+      {/* Industry Selection */}
+      <div style={{ marginBottom: '20px' }}>
+        <label style={{ display: 'block', marginBottom: '8px', fontWeight: '600', color: '#555' }}>
+          Industry
+        </label>
+        <select
+          value={state.industry}
+          onChange={handleIndustryChange}
+          style={{
+            width: '100%',
+            padding: '12px',
+            border: '1px solid #ddd',
+            borderRadius: '8px',
+            fontSize: '14px',
+            fontFamily: 'inherit',
+            backgroundColor: 'white',
+            cursor: 'pointer',
+          }}
+        >
+          {industries.map(ind => (
+            <option key={ind} value={ind}>{ind}</option>
+          ))}
+        </select>
+      </div>
+
+      {/* Target Audience */}
+      <div style={{ marginBottom: '20px' }}>
+        <label style={{ display: 'block', marginBottom: '8px', fontWeight: '600', color: '#555' }}>
+          Target Audience
+        </label>
+        <input
+          type="text"
+          value={state.targetAudience}
+          onChange={handleAudienceChange}
+          placeholder="e.g., Digital marketers, Small business owners"
+          style={{
+            width: '100%',
+            padding: '12px',
+            border: '1px solid #ddd',
+            borderRadius: '8px',
+            fontSize: '14px',
+            fontFamily: 'inherit',
+            boxSizing: 'border-box',
+          }}
+        />
+      </div>
+
+      <div style={{
+        padding: '12px',
+        backgroundColor: '#f0f7ff',
+        borderRadius: '8px',
+        border: '1px solid #b3d9ff',
+        fontSize: '13px',
+        color: '#004085',
+      }}>
+        💡 <strong>Tip:</strong> Be specific with your keywords. The more precise your keywords, the better your research results.
+      </div>
+    </div>
+  );
+};
+
diff --git a/frontend/src/components/Research/steps/StepOptions.tsx b/frontend/src/components/Research/steps/StepOptions.tsx
new file mode 100644
index 00000000..f9898566
--- /dev/null
+++ b/frontend/src/components/Research/steps/StepOptions.tsx
@@ -0,0 +1,182 @@
+import React from 'react';
+import { WizardStepProps, ModeCardInfo } from '../types/research.types';
+import { ResearchProvider } from '../../../services/blogWriterApi';
+
+const modeCards: ModeCardInfo[] = [
+  {
+    mode: 'basic',
+    title: 'Basic Research',
+    description: 'Quick keyword-focused analysis for fast results',
+    features: [
+      'Primary & secondary keywords',
+      'Current trends overview',
+      'Top 5 content angles',
+      'Key statistics',
+    ],
+    icon: '⚡',
+  },
+  {
+    mode: 'comprehensive',
+    title: 'Comprehensive Research',
+    description: 'Deep analysis with full competitive intelligence',
+    features: [
+      'All basic features',
+      'Expert quotes & opinions',
+      'Competitor analysis',
+      'Market forecasts',
+      'Best practices & case studies',
+      'Content gaps identification',
+    ],
+    icon: '📊',
+  },
+  {
+    mode: 'targeted',
+    title: 'Targeted Research',
+    description: 'Customize what you need most',
+    features: [
+      'Select specific components',
+      'Choose date ranges',
+      'Filter source types',
+      'Control depth & scope',
+    ],
+    icon: '🎯',
+  },
+];
+
+export const StepOptions: React.FC<WizardStepProps> = ({ state, onUpdate }) => {
+  const handleModeChange = (mode: any) => {
+    onUpdate({ researchMode: mode });
+  };
+
+  const handleProviderChange = (provider: ResearchProvider) => {
+    onUpdate({ config: { ...state.config, provider } });
+  };
+
+  return (
+    <div style={{ padding: '24px', maxWidth: '1000px', margin: '0 auto' }}>
+      <h2 style={{ marginBottom: '8px', color: '#333' }}>Choose Research Mode</h2>
+      <p style={{ marginBottom: '24px', color: '#666', fontSize: '15px' }}>
+        Select the type of research that best fits your needs.
+      </p>
+
+      <div style={{
+        display: 'grid',
+        gridTemplateColumns: 'repeat(auto-fit, minmax(280px, 1fr))',
+        gap: '20px',
+        marginBottom: '24px',
+      }}>
+        {modeCards.map(card => (
+          <div
+            key={card.mode}
+            onClick={() => handleModeChange(card.mode)}
+            style={{
+              border: state.researchMode === card.mode ? '2px solid #1976d2' : '2px solid #e0e0e0',
+              borderRadius: '12px',
+              padding: '24px',
+              cursor: 'pointer',
+              transition: 'all 0.2s ease',
+              backgroundColor: state.researchMode === card.mode ? '#f0f7ff' : 'white',
+            }}
+            onMouseEnter={(e) => {
+              if (state.researchMode !== card.mode) {
+                e.currentTarget.style.borderColor = '#90caf9';
+                e.currentTarget.style.backgroundColor = '#fafafa';
+              }
+            }}
+            onMouseLeave={(e) => {
+              if (state.researchMode !== card.mode) {
+                e.currentTarget.style.borderColor = '#e0e0e0';
+                e.currentTarget.style.backgroundColor = 'white';
+              }
+            }}
+          >
+            <div style={{ display: 'flex', alignItems: 'center', marginBottom: '12px' }}>
+              <span style={{ fontSize: '32px', marginRight: '12px' }}>{card.icon}</span>
+              <h3 style={{ margin: 0, color: '#333', fontSize: '18px' }}>{card.title}</h3>
+            </div>
+            <p style={{ marginBottom: '16px', color: '#666', fontSize: '14px' }}>
+              {card.description}
+            </p>
+            <ul style={{ margin: 0, paddingLeft: '20px', fontSize: '13px', color: '#555' }}>
+              {card.features.map((feature, idx) => (
+                <li key={idx} style={{ marginBottom: '4px' }}>{feature}</li>
+              ))}
+            </ul>
+            {state.researchMode === card.mode && (
+              <div style={{
+                marginTop: '16px',
+                padding: '8px',
+                backgroundColor: '#1976d2',
+                color: 'white',
+                borderRadius: '6px',
+                textAlign: 'center',
+                fontSize: '13px',
+                fontWeight: '600',
+              }}>
+                ✓ Selected
+              </div>
+            )}
+          </div>
+        ))}
+      </div>
+
+      {state.researchMode !== 'basic' && (
+        <div style={{ marginBottom: '24px', border: '1px solid #e0e0e0', borderRadius: '8px', padding: '15px', backgroundColor: '#f9f9f9' }}>
+          <h3 style={{ marginTop: 0, marginBottom: '12px', color: '#555', fontSize: '16px' }}>
+            🔍 Research Provider
+          </h3>
+          <div style={{ display: 'flex', gap: '12px' }}>
+            <div
+              onClick={() => handleProviderChange('google')}
+              style={{
+                flex: 1,
+                padding: '12px',
+                border: '2px solid',
+                borderColor: (state.config.provider === 'google' || !state.config.provider) ? '#1976d2' : '#ddd',
+                backgroundColor: (state.config.provider === 'google' || !state.config.provider) ? '#e3f2fd' : 'white',
+                borderRadius: '6px',
+                cursor: 'pointer',
+                transition: 'all 0.2s',
+              }}
+            >
+              <div style={{ fontWeight: '600', marginBottom: '4px' }}>Google Search</div>
+              <div style={{ fontSize: '11px', color: '#666' }}>
+                Fast, broad coverage, trending topics
+              </div>
+            </div>
+            <div
+              onClick={() => handleProviderChange('exa')}
+              style={{
+                flex: 1,
+                padding: '12px',
+                border: '2px solid',
+                borderColor: state.config.provider === 'exa' ? '#7c3aed' : '#ddd',
+                backgroundColor: state.config.provider === 'exa' ? '#f3e8ff' : 'white',
+                borderRadius: '6px',
+                cursor: 'pointer',
+                transition: 'all 0.2s',
+              }}
+            >
+              <div style={{ fontWeight: '600', marginBottom: '4px' }}>Exa Neural</div>
+              <div style={{ fontSize: '11px', color: '#666' }}>
+                Deep research, rich citations, semantic search
+              </div>
+            </div>
+          </div>
+        </div>
+      )}
+
+      <div style={{
+        padding: '12px',
+        backgroundColor: '#fff3e0',
+        borderRadius: '8px',
+        border: '1px solid #ffb74d',
+        fontSize: '13px',
+        color: '#e65100',
+      }}>
+        ℹ️ <strong>Note:</strong> You can always run additional research if you need more information later.
+      </div>
+    </div>
+  );
+};
+
diff --git a/frontend/src/components/Research/steps/StepProgress.tsx b/frontend/src/components/Research/steps/StepProgress.tsx
new file mode 100644
index 00000000..b4315b10
--- /dev/null
+++ b/frontend/src/components/Research/steps/StepProgress.tsx
@@ -0,0 +1,153 @@
+import React, { useEffect } from 'react';
+import { WizardStepProps } from '../types/research.types';
+import { useResearchExecution } from '../hooks/useResearchExecution';
+
+export const StepProgress: React.FC<WizardStepProps> = ({ state, onNext, onUpdate }) => {
+  const { executeResearch, stopExecution, isExecuting, error, progressMessages, currentStatus } = useResearchExecution();
+
+  useEffect(() => {
+    // Start research when this step is reached
+    const startResearch = async () => {
+      const taskId = await executeResearch(state);
+      if (taskId === 'cached') {
+        // If cached, move to results immediately
+        // The parent will handle this
+      }
+    };
+
+    startResearch();
+
+    return () => {
+      if (isExecuting) {
+        stopExecution();
+      }
+    };
+  }, []); // Run once on mount
+
+  // Move to next step when research completes
+  useEffect(() => {
+    if (!isExecuting && progressMessages.length > 0) {
+      // Small delay to show final message
+      const timer = setTimeout(() => {
+        onNext();
+      }, 1000);
+      return () => clearTimeout(timer);
+    }
+  }, [isExecuting, progressMessages.length, onNext]);
+
+  const getStatusIcon = () => {
+    if (error) return '❌';
+    if (!isExecuting && progressMessages.length > 0) return '✅';
+    if (currentStatus === 'completed') return '✅';
+    return '🔄';
+  };
+
+  const getStatusColor = () => {
+    if (error) return '#f44336';
+    if (!isExecuting && progressMessages.length > 0) return '#4caf50';
+    return '#1976d2';
+  };
+
+  return (
+    <div style={{ padding: '24px', maxWidth: '800px', margin: '0 auto' }}>
+      <h2 style={{ marginBottom: '8px', color: '#333' }}>Researching...</h2>
+      <p style={{ marginBottom: '24px', color: '#666', fontSize: '15px' }}>
+        Gathering insights from Google Search grounding
+      </p>
+
+      {/* Status Display */}
+      <div style={{
+        backgroundColor: '#f5f5f5',
+        borderRadius: '12px',
+        padding: '24px',
+        marginBottom: '24px',
+        textAlign: 'center',
+      }}>
+        <div style={{ fontSize: '48px', marginBottom: '16px' }}>{getStatusIcon()}</div>
+        
+        {error ? (
+          <>
+            <h3 style={{ color: getStatusColor(), marginBottom: '8px' }}>Error</h3>
+            <p style={{ color: '#666', fontSize: '14px' }}>{error}</p>
+            <button
+              onClick={() => window.location.reload()}
+              style={{
+                marginTop: '16px',
+                padding: '8px 16px',
+                backgroundColor: '#1976d2',
+                color: 'white',
+                border: 'none',
+                borderRadius: '6px',
+                cursor: 'pointer',
+                fontSize: '14px',
+              }}
+            >
+              Retry
+            </button>
+          </>
+        ) : (
+          <>
+            <h3 style={{ color: getStatusColor(), marginBottom: '8px' }}>
+              {currentStatus === 'completed' ? 'Complete!' : 'In Progress'}
+            </h3>
+            <p style={{ color: '#666', fontSize: '14px' }}>
+              {isExecuting ? 'Analyzing sources and generating insights...' : 'Finalizing results...'}
+            </p>
+          </>
+        )}
+      </div>
+
+      {/* Progress Messages */}
+      {progressMessages.length > 0 && (
+        <div style={{
+          backgroundColor: 'white',
+          border: '1px solid #e0e0e0',
+          borderRadius: '8px',
+          maxHeight: '300px',
+          overflow: 'auto',
+        }}>
+          <div style={{ padding: '16px', borderBottom: '1px solid #e0e0e0' }}>
+            <strong style={{ fontSize: '14px', color: '#333' }}>Progress Updates</strong>
+          </div>
+          {progressMessages.map((msg, idx) => (
+            <div
+              key={idx}
+              style={{
+                padding: '12px 16px',
+                borderBottom: idx < progressMessages.length - 1 ? '1px solid #f0f0f0' : 'none',
+                fontSize: '13px',
+                color: '#555',
+              }}
+            >
+              {idx === progressMessages.length - 1 && isExecuting && (
+                <span style={{ marginRight: '8px' }}>🔄</span>
+              )}
+              {msg.message}
+            </div>
+          ))}
+        </div>
+      )}
+
+      {/* Cancel Button */}
+      {isExecuting && (
+        <div style={{ marginTop: '24px', textAlign: 'center' }}>
+          <button
+            onClick={stopExecution}
+            style={{
+              padding: '8px 16px',
+              backgroundColor: '#f44336',
+              color: 'white',
+              border: 'none',
+              borderRadius: '6px',
+              cursor: 'pointer',
+              fontSize: '14px',
+            }}
+          >
+            Cancel Research
+          </button>
+        </div>
+      )}
+    </div>
+  );
+};
+
diff --git a/frontend/src/components/Research/steps/StepResults.tsx b/frontend/src/components/Research/steps/StepResults.tsx
new file mode 100644
index 00000000..4c1f8431
--- /dev/null
+++ b/frontend/src/components/Research/steps/StepResults.tsx
@@ -0,0 +1,103 @@
+import React from 'react';
+import { WizardStepProps } from '../types/research.types';
+import { ResearchResults } from '../../BlogWriter/ResearchResults';
+import { BlogResearchResponse } from '../../../services/blogWriterApi';
+
+export const StepResults: React.FC<WizardStepProps> = ({ state, onBack }) => {
+  if (!state.results) {
+    return (
+      <div style={{ padding: '24px', textAlign: 'center' }}>
+        <p style={{ color: '#666' }}>No results available</p>
+      </div>
+    );
+  }
+
+  const handleExport = () => {
+    const dataStr = JSON.stringify(state.results, null, 2);
+    const dataBlob = new Blob([dataStr], { type: 'application/json' });
+    const url = URL.createObjectURL(dataBlob);
+    const link = document.createElement('a');
+    link.href = url;
+    link.download = `research-${state.keywords.join('-')}-${Date.now()}.json`;
+    link.click();
+    URL.revokeObjectURL(url);
+  };
+
+  return (
+    <div style={{ maxWidth: '1200px', margin: '0 auto', padding: '24px' }}>
+      <div style={{
+        display: 'flex',
+        justifyContent: 'space-between',
+        alignItems: 'center',
+        marginBottom: '24px',
+        flexWrap: 'wrap',
+        gap: '16px',
+      }}>
+        <h2 style={{ margin: 0, color: '#333' }}>Research Results</h2>
+        
+        <div style={{ display: 'flex', gap: '8px', flexWrap: 'wrap' }}>
+          <button
+            onClick={handleExport}
+            style={{
+              padding: '8px 16px',
+              backgroundColor: '#1976d2',
+              color: 'white',
+              border: 'none',
+              borderRadius: '6px',
+              cursor: 'pointer',
+              fontSize: '14px',
+              display: 'flex',
+              alignItems: 'center',
+              gap: '6px',
+            }}
+          >
+            📥 Export JSON
+          </button>
+          
+          <button
+            onClick={onBack}
+            style={{
+              padding: '8px 16px',
+              backgroundColor: '#f5f5f5',
+              color: '#333',
+              border: '1px solid #ddd',
+              borderRadius: '6px',
+              cursor: 'pointer',
+              fontSize: '14px',
+            }}
+          >
+            ← Start New Research
+          </button>
+        </div>
+      </div>
+
+      {/* Results Display */}
+      <div style={{
+        backgroundColor: 'white',
+        borderRadius: '8px',
+        border: '1px solid #e0e0e0',
+        overflow: 'hidden',
+      }}>
+        <ResearchResults research={state.results} />
+      </div>
+
+      {/* Action Section */}
+      <div style={{
+        marginTop: '24px',
+        padding: '16px',
+        backgroundColor: '#f0f7ff',
+        borderRadius: '8px',
+        border: '1px solid #b3d9ff',
+      }}>
+        <h4 style={{ marginBottom: '8px', color: '#004085' }}>Next Steps</h4>
+        <ul style={{ margin: 0, paddingLeft: '20px', color: '#004085', fontSize: '14px' }}>
+          <li>Review the research insights and sources</li>
+          <li>Explore content angles and competitor analysis</li>
+          <li>Use this research to create your blog outline</li>
+          <li>Export the data for reference</li>
+        </ul>
+      </div>
+    </div>
+  );
+};
+
diff --git a/frontend/src/components/Research/types/research.types.ts b/frontend/src/components/Research/types/research.types.ts
new file mode 100644
index 00000000..63f674be
--- /dev/null
+++ b/frontend/src/components/Research/types/research.types.ts
@@ -0,0 +1,34 @@
+import { BlogResearchResponse, ResearchMode, ResearchConfig } from '../../../services/blogWriterApi';
+
+export interface WizardState {
+  currentStep: number;
+  keywords: string[];
+  industry: string;
+  targetAudience: string;
+  researchMode: ResearchMode;
+  config: ResearchConfig;
+  results: BlogResearchResponse | null;
+}
+
+export interface WizardStepProps {
+  state: WizardState;
+  onUpdate: (updates: Partial<WizardState>) => void;
+  onNext: () => void;
+  onBack: () => void;
+}
+
+export interface ResearchWizardProps {
+  onComplete?: (results: BlogResearchResponse) => void;
+  onCancel?: () => void;
+  initialKeywords?: string[];
+  initialIndustry?: string;
+}
+
+export interface ModeCardInfo {
+  mode: ResearchMode;
+  title: string;
+  description: string;
+  features: string[];
+  icon: string;
+}
+
diff --git a/frontend/src/components/Research/utils/researchUtils.ts b/frontend/src/components/Research/utils/researchUtils.ts
new file mode 100644
index 00000000..c569620f
--- /dev/null
+++ b/frontend/src/components/Research/utils/researchUtils.ts
@@ -0,0 +1,17 @@
+// Utility functions for research component
+
+export const formatKeywords = (keywords: string[]): string => {
+  return keywords.join(', ');
+};
+
+export const parseKeywords = (keywordsString: string): string[] => {
+  return keywordsString
+    .split(',')
+    .map(k => k.trim())
+    .filter(Boolean);
+};
+
+export const validateKeywords = (keywords: string[]): boolean => {
+  return keywords.length > 0 && keywords.every(k => k.trim().length > 0);
+};
+
diff --git a/frontend/src/components/SubscriptionExpiredModal.tsx b/frontend/src/components/SubscriptionExpiredModal.tsx
index 620b7064..4e0dbeb8 100644
--- a/frontend/src/components/SubscriptionExpiredModal.tsx
+++ b/frontend/src/components/SubscriptionExpiredModal.tsx
@@ -41,12 +41,20 @@ const SubscriptionExpiredModal: React.FC<SubscriptionExpiredModalProps> = ({
 }) => {
   // Debug logging to verify modal state
   React.useEffect(() => {
+    console.log('SubscriptionExpiredModal: State update', {
+      open,
+      errorData,
+      hasUsageInfo: !!errorData?.usage_info,
+      errorDataKeys: errorData ? Object.keys(errorData) : null
+    });
     if (open) {
-      console.log('SubscriptionExpiredModal: Modal opened', {
+      console.log('SubscriptionExpiredModal: Modal should be visible now', {
         open,
         errorData,
         hasUsageInfo: !!errorData?.usage_info
       });
+    } else {
+      console.log('SubscriptionExpiredModal: Modal is closed');
     }
   }, [open, errorData]);
   
diff --git a/frontend/src/components/WixCallbackPage/WixCallbackPage.tsx b/frontend/src/components/WixCallbackPage/WixCallbackPage.tsx
index a4d2a5db..af0ba615 100644
--- a/frontend/src/components/WixCallbackPage/WixCallbackPage.tsx
+++ b/frontend/src/components/WixCallbackPage/WixCallbackPage.tsx
@@ -57,8 +57,18 @@ const WixCallbackPage: React.FC = () => {
             return;
           }
         } catch {}
-        // Fallback redirect for same-tab flow and let onboarding hook mark Wix as connected
-        window.location.replace('/onboarding?step=5&wix_connected=true');
+        // Fallback redirect for same-tab flow - check if we have a stored redirect URL
+        const redirectUrl = sessionStorage.getItem('wix_oauth_redirect');
+        console.log('[Wix Callback] Checking redirect URL:', redirectUrl);
+        if (redirectUrl) {
+          console.log('[Wix Callback] Redirecting to stored URL:', redirectUrl);
+          sessionStorage.removeItem('wix_oauth_redirect');
+          window.location.replace(redirectUrl);
+        } else {
+          // Default to onboarding if no redirect URL stored
+          console.warn('[Wix Callback] No redirect URL found, defaulting to onboarding');
+          window.location.replace('/onboarding?step=5&wix_connected=true');
+        }
       } catch (e: any) {
         setError(e?.message || 'OAuth callback failed');
         try {
diff --git a/frontend/src/components/shared/CopilotKitDegradedBanner.tsx b/frontend/src/components/shared/CopilotKitDegradedBanner.tsx
new file mode 100644
index 00000000..97c9bdab
--- /dev/null
+++ b/frontend/src/components/shared/CopilotKitDegradedBanner.tsx
@@ -0,0 +1,74 @@
+import React, { useState } from 'react';
+import { Alert, AlertTitle, IconButton, Collapse } from '@mui/material';
+import { Close as CloseIcon, Warning as WarningIcon } from '@mui/icons-material';
+import { useCopilotKitHealth } from '../../hooks/useCopilotKitHealth';
+
+interface CopilotKitDegradedBannerProps {
+  /**
+   * Position of the banner
+   * @default 'top'
+   */
+  position?: 'top' | 'bottom';
+  /**
+   * Whether the banner is dismissible
+   * @default true
+   */
+  dismissible?: boolean;
+}
+
+/**
+ * Banner component that displays when CopilotKit is unavailable
+ * Non-intrusive notification that chat is unavailable but app continues to work
+ */
+export const CopilotKitDegradedBanner: React.FC<CopilotKitDegradedBannerProps> = ({
+  position = 'top',
+  dismissible = true,
+}) => {
+  const { isAvailable, errorMessage, isChecking } = useCopilotKitHealth();
+  const [dismissed, setDismissed] = useState(false);
+
+  // Don't show if CopilotKit is available, checking, or dismissed
+  if (isAvailable || isChecking || dismissed) {
+    return null;
+  }
+
+  const handleDismiss = () => {
+    setDismissed(true);
+  };
+
+  return (
+    <Collapse in={!dismissed}>
+      <Alert
+        severity="warning"
+        icon={<WarningIcon />}
+        action={
+          dismissible ? (
+            <IconButton
+              aria-label="close"
+              color="inherit"
+              size="small"
+              onClick={handleDismiss}
+            >
+              <CloseIcon fontSize="inherit" />
+            </IconButton>
+          ) : null
+        }
+        sx={{
+          position: 'fixed',
+          [position]: 0,
+          left: 0,
+          right: 0,
+          zIndex: 1300, // Above most content but below modals
+          borderRadius: 0, // Full width banner
+          boxShadow: 2,
+        }}
+      >
+        <AlertTitle>Chat Unavailable</AlertTitle>
+        {errorMessage || 'CopilotKit service is currently unavailable. You can still use all features with manual controls.'}
+      </Alert>
+    </Collapse>
+  );
+};
+
+export default CopilotKitDegradedBanner;
+
diff --git a/frontend/src/contexts/CopilotKitHealthContext.tsx b/frontend/src/contexts/CopilotKitHealthContext.tsx
new file mode 100644
index 00000000..5aa0679c
--- /dev/null
+++ b/frontend/src/contexts/CopilotKitHealthContext.tsx
@@ -0,0 +1,157 @@
+import React, { createContext, useContext, useState, useCallback, ReactNode } from 'react';
+
+interface CopilotKitHealthState {
+  isHealthy: boolean;
+  isChecking: boolean;
+  lastChecked: Date | null;
+  errorMessage: string | null;
+  retryCount: number;
+  isAvailable: boolean; // Alias for isHealthy, for clearer semantics
+}
+
+interface CopilotKitHealthContextType extends CopilotKitHealthState {
+  checkHealth: () => Promise<void>;
+  markUnhealthy: (errorMessage?: string) => void;
+  markHealthy: () => void;
+  resetHealth: () => void;
+}
+
+const CopilotKitHealthContext = createContext<CopilotKitHealthContextType | undefined>(undefined);
+
+export const useCopilotKitHealthContext = () => {
+  const context = useContext(CopilotKitHealthContext);
+  if (!context) {
+    throw new Error('useCopilotKitHealthContext must be used within CopilotKitHealthProvider');
+  }
+  return context;
+};
+
+interface CopilotKitHealthProviderProps {
+  children: ReactNode;
+  initialHealthStatus?: boolean;
+}
+
+export const CopilotKitHealthProvider: React.FC<CopilotKitHealthProviderProps> = ({
+  children,
+  initialHealthStatus = true,
+}) => {
+  const [state, setState] = useState<CopilotKitHealthState>({
+    isHealthy: initialHealthStatus,
+    isChecking: false,
+    lastChecked: null,
+    errorMessage: null,
+    retryCount: 0,
+    isAvailable: initialHealthStatus,
+  });
+
+  const markHealthy = useCallback(() => {
+    setState((prev) => ({
+      ...prev,
+      isHealthy: true,
+      isAvailable: true,
+      errorMessage: null,
+      retryCount: 0,
+      lastChecked: new Date(),
+    }));
+  }, []);
+
+  const markUnhealthy = useCallback((errorMessage?: string) => {
+    setState((prev) => ({
+      ...prev,
+      isHealthy: false,
+      isAvailable: false,
+      errorMessage: errorMessage || 'CopilotKit is unavailable',
+      lastChecked: new Date(),
+      retryCount: prev.retryCount + 1,
+    }));
+  }, []);
+
+  // Listen for CopilotKit error events from App.tsx
+  React.useEffect(() => {
+    const handleCopilotKitError = (event: Event) => {
+      const customEvent = event as CustomEvent;
+      const { errorMessage, isFatal } = customEvent.detail || {};
+      
+      if (isFatal) {
+        markUnhealthy(errorMessage || 'CopilotKit fatal error');
+      } else {
+        // For transient errors, just log but don't mark as unhealthy immediately
+        // Let the health check determine if it's truly down
+        console.warn('CopilotKit transient error:', errorMessage);
+      }
+    };
+
+    window.addEventListener('copilotkit-error', handleCopilotKitError as EventListener);
+    return () => {
+      window.removeEventListener('copilotkit-error', handleCopilotKitError as EventListener);
+    };
+  }, [markUnhealthy]);
+
+  const checkHealth = useCallback(async () => {
+    setState((prev) => ({ ...prev, isChecking: true }));
+
+    try {
+      // Try to check CopilotKit status endpoint
+      // This is a lightweight check that doesn't require full CopilotKit initialization
+      const response = await fetch('https://api.cloud.copilotkit.ai/ciu', {
+        method: 'GET',
+        headers: {
+          'x-copilotcloud-public-api-key': process.env.REACT_APP_COPILOTKIT_PUBLIC_API_KEY || '',
+        },
+        // Use a short timeout to avoid blocking
+        signal: AbortSignal.timeout(3000),
+      });
+
+      if (response.ok) {
+        markHealthy();
+      } else {
+        markUnhealthy(`CopilotKit status check failed: ${response.status}`);
+      }
+    } catch (error: any) {
+      // Handle various error types
+      let errorMsg = 'CopilotKit health check failed';
+      
+      if (error.name === 'AbortError' || error.name === 'TimeoutError') {
+        errorMsg = 'CopilotKit health check timed out';
+      } else if (error.message?.includes('CORS')) {
+        errorMsg = 'CopilotKit CORS error - service may be unavailable';
+      } else if (error.message?.includes('certificate') || error.message?.includes('SSL')) {
+        errorMsg = 'CopilotKit SSL certificate error';
+      } else if (error.message?.includes('network') || error.message?.includes('Failed to fetch')) {
+        errorMsg = 'CopilotKit network error - service may be down';
+      } else {
+        errorMsg = error.message || 'Unknown error checking CopilotKit health';
+      }
+
+      markUnhealthy(errorMsg);
+    } finally {
+      setState((prev) => ({ ...prev, isChecking: false }));
+    }
+  }, [markHealthy, markUnhealthy]);
+
+  const resetHealth = useCallback(() => {
+    setState({
+      isHealthy: initialHealthStatus,
+      isChecking: false,
+      lastChecked: null,
+      errorMessage: null,
+      retryCount: 0,
+      isAvailable: initialHealthStatus,
+    });
+  }, [initialHealthStatus]);
+
+  const value: CopilotKitHealthContextType = {
+    ...state,
+    checkHealth,
+    markUnhealthy,
+    markHealthy,
+    resetHealth,
+  };
+
+  return (
+    <CopilotKitHealthContext.Provider value={value}>
+      {children}
+    </CopilotKitHealthContext.Provider>
+  );
+};
+
diff --git a/frontend/src/contexts/SubscriptionContext.tsx b/frontend/src/contexts/SubscriptionContext.tsx
index c358c516..cefc713b 100644
--- a/frontend/src/contexts/SubscriptionContext.tsx
+++ b/frontend/src/contexts/SubscriptionContext.tsx
@@ -1,6 +1,7 @@
 import React, { createContext, useContext, useState, useEffect, ReactNode, useCallback, useRef } from 'react';
 import { apiClient, setGlobalSubscriptionErrorHandler } from '../api/client';
 import SubscriptionExpiredModal from '../components/SubscriptionExpiredModal';
+import { saveNavigationState, getCurrentPhaseForTool } from '../utils/navigationState';
 
 export interface SubscriptionLimits {
   gemini_calls: number;
@@ -221,11 +222,29 @@ export const SubscriptionProvider: React.FC<SubscriptionProviderProps> = ({ chil
   }, []);
 
   const handleRenewSubscription = useCallback(() => {
-    // Save current location so we can return after renewal
+    // Save current location and phase so we can return after renewal
     const currentPath = window.location.pathname;
-    sessionStorage.setItem('subscription_referrer', currentPath);
     
-    console.log('SubscriptionContext: Navigating to pricing page, saved referrer:', currentPath);
+    // Detect tool from path
+    let tool: string | undefined;
+    if (currentPath.includes('/blog-writer') || currentPath.includes('/blogwriter')) {
+      tool = 'blog-writer';
+    }
+    
+    // Get current phase for the tool if applicable
+    let phase: string | null = null;
+    if (tool) {
+      phase = getCurrentPhaseForTool(tool);
+    }
+    
+    // Save navigation state (path, phase, tool)
+    saveNavigationState(currentPath, phase || undefined, tool);
+    
+    console.log('SubscriptionContext: Navigating to pricing page, saved navigation state:', {
+      path: currentPath,
+      phase,
+      tool
+    });
     window.location.href = '/pricing';
   }, []);
 
@@ -258,13 +277,30 @@ export const SubscriptionProvider: React.FC<SubscriptionProviderProps> = ({ chil
         errorData = errorData[0] || {};
       }
       
-      // Check for usage_info in various possible locations
+      // CRITICAL: FastAPI wraps HTTPException detail in a 'detail' field
+      // If errorData has a 'detail' field, extract it (this is the actual error data)
+      if (errorData.detail && typeof errorData.detail === 'object') {
+        console.log('SubscriptionContext: Found FastAPI detail wrapper, extracting detail field');
+        errorData = errorData.detail;
+      }
+      
+      // Check for usage_info in various possible locations (now that we've unwrapped FastAPI detail)
       const usageInfo = errorData.usage_info || 
                        (errorData.current_calls !== undefined ? errorData : null) ||
+                       (errorData.requested_tokens !== undefined ? errorData : null) ||
+                       (errorData.current_tokens !== undefined ? errorData : null) ||
                        null;
       
-      // Usage limit error: 429 status with usage info OR 429 status without explicit expiration
-      const isUsageLimitError = status === 429 && (usageInfo || errorData.provider || errorData.message);
+      // Usage limit error: 429 status with usage info OR provider OR message indicating token/call limits
+      const hasUsageIndicators = usageInfo || 
+                                 errorData.provider || 
+                                 errorData.message?.includes('limit') ||
+                                 errorData.error?.includes('limit') ||
+                                 errorData.requested_tokens !== undefined ||
+                                 errorData.current_tokens !== undefined ||
+                                 errorData.current_calls !== undefined;
+      
+      const isUsageLimitError = status === 429 && hasUsageIndicators;
       const isSubscriptionExpired = status === 402 || (status === 429 && !isUsageLimitError);
       
       console.log('SubscriptionContext: Error analysis', {
@@ -280,16 +316,30 @@ export const SubscriptionProvider: React.FC<SubscriptionProviderProps> = ({ chil
       // For usage limit errors (429 with usage_info), always show modal - even for active subscriptions
       // Ignore grace window and cooldown for usage limit errors (user needs to know immediately)
       if (isUsageLimitError) {
+        // Build usage_info from various possible locations
+        const finalUsageInfo = usageInfo || 
+                              (errorData.requested_tokens !== undefined ? {
+                                provider: errorData.provider,
+                                current_tokens: errorData.current_tokens,
+                                requested_tokens: errorData.requested_tokens,
+                                limit: errorData.limit,
+                                type: 'tokens',
+                                ...errorData
+                              } : null) ||
+                              errorData;
+        
         const modalData = {
           provider: errorData.provider || usageInfo?.provider || 'unknown',
-          usage_info: usageInfo || errorData,
+          usage_info: finalUsageInfo || errorData,
           message: errorData.message || errorData.error || 'You have reached your usage limit.'
         };
         
         console.log('SubscriptionContext: Usage limit exceeded, showing modal (ignoring grace window/cooldown)', {
           modalData,
           errorData: Object.keys(errorData),
-          usageInfo: usageInfo ? Object.keys(usageInfo) : null
+          usageInfo: usageInfo ? Object.keys(usageInfo) : null,
+          currentShowModal: showModal,
+          currentModalErrorData: modalErrorData
         });
         
         // Set flag to mark this as a usage limit modal (should never be auto-closed)
@@ -298,7 +348,17 @@ export const SubscriptionProvider: React.FC<SubscriptionProviderProps> = ({ chil
         setShowModal(true);
         setLastModalShowTime(now);
         
-        console.log('SubscriptionContext: Modal state updated - showModal should be true, isUsageLimitModal = true');
+        console.log('SubscriptionContext: Modal state updated - showModal should be true, isUsageLimitModal = true', {
+          showModal: true,
+          isUsageLimitModal: true,
+          modalErrorData: modalData
+        });
+        
+        // Force a re-render check
+        setTimeout(() => {
+          console.log('SubscriptionContext: State check after timeout - showModal:', showModal, 'modalErrorData:', modalErrorData);
+        }, 100);
+        
         return true;
       }
       
diff --git a/frontend/src/hooks/useBlogWriterState.ts b/frontend/src/hooks/useBlogWriterState.ts
index 73b06356..27d001a6 100644
--- a/frontend/src/hooks/useBlogWriterState.ts
+++ b/frontend/src/hooks/useBlogWriterState.ts
@@ -116,10 +116,10 @@ export const useBlogWriterState = () => {
         }
       }
       
-      // Save to localStorage for persistence
+      // Save to localStorage for persistence (using shared cache utility)
       try {
-        localStorage.setItem('blog_outline', JSON.stringify(result.outline));
-        localStorage.setItem('blog_title_options', JSON.stringify(result.title_options || []));
+        const { blogWriterCache } = require('../services/blogWriterCache');
+        blogWriterCache.cacheOutline(result.outline, result.title_options);
         localStorage.setItem('blog_selected_title', result.title_options?.[0] || '');
         console.log('Saved outline data to localStorage');
       } catch (error) {
diff --git a/frontend/src/hooks/useCopilotKitHealth.ts b/frontend/src/hooks/useCopilotKitHealth.ts
new file mode 100644
index 00000000..b6259c3d
--- /dev/null
+++ b/frontend/src/hooks/useCopilotKitHealth.ts
@@ -0,0 +1,161 @@
+import { useEffect, useRef, useCallback } from 'react';
+import { useCopilotKitHealthContext } from '../contexts/CopilotKitHealthContext';
+
+interface UseCopilotKitHealthOptions {
+  /**
+   * Initial delay before first health check (milliseconds)
+   * @default 1000
+   */
+  initialDelay?: number;
+  /**
+   * Interval between health checks when healthy (milliseconds)
+   * @default 60000 (1 minute)
+   */
+  healthyInterval?: number;
+  /**
+   * Exponential backoff intervals when unhealthy (milliseconds)
+   * @default [5000, 10000, 30000, 60000]
+   */
+  unhealthyIntervals?: number[];
+  /**
+   * Enable automatic health checking
+   * @default true
+   */
+  enabled?: boolean;
+}
+
+/**
+ * Hook to monitor CopilotKit health status with automatic polling
+ * Uses exponential backoff when unhealthy
+ */
+export const useCopilotKitHealth = (options: UseCopilotKitHealthOptions = {}) => {
+  const {
+    initialDelay = 1000,
+    healthyInterval = 60000, // 1 minute
+    unhealthyIntervals = [5000, 10000, 30000, 60000], // 5s, 10s, 30s, 60s
+    enabled = true,
+  } = options;
+
+  const {
+    isHealthy,
+    isChecking,
+    lastChecked,
+    errorMessage,
+    retryCount,
+    isAvailable,
+    checkHealth,
+    markUnhealthy,
+  } = useCopilotKitHealthContext();
+
+  const intervalRef = useRef<NodeJS.Timeout | null>(null);
+  const timeoutRef = useRef<NodeJS.Timeout | null>(null);
+
+  const scheduleNextCheck = useCallback(() => {
+    // Clear any existing timeouts/intervals
+    if (intervalRef.current) {
+      clearInterval(intervalRef.current);
+      intervalRef.current = null;
+    }
+    if (timeoutRef.current) {
+      clearTimeout(timeoutRef.current);
+      timeoutRef.current = null;
+    }
+
+    if (!enabled) return;
+
+    // Calculate next check interval based on health status
+    let nextInterval: number;
+    
+    if (isHealthy) {
+      // When healthy, use standard interval
+      nextInterval = healthyInterval;
+    } else {
+      // When unhealthy, use exponential backoff
+      const intervalIndex = Math.min(retryCount, unhealthyIntervals.length - 1);
+      nextInterval = unhealthyIntervals[intervalIndex];
+    }
+
+    // Schedule next check
+    timeoutRef.current = setTimeout(() => {
+      checkHealth();
+    }, nextInterval);
+  }, [enabled, isHealthy, retryCount, healthyInterval, unhealthyIntervals, checkHealth]);
+
+  // Initial health check on mount
+  useEffect(() => {
+    if (!enabled) return;
+
+    // Initial delay before first check
+    const initialTimeout = setTimeout(() => {
+      checkHealth();
+    }, initialDelay);
+
+    return () => {
+      clearTimeout(initialTimeout);
+    };
+  }, [enabled, initialDelay, checkHealth]);
+
+  // Schedule next check after health status changes
+  useEffect(() => {
+    if (!enabled || isChecking) return;
+
+    scheduleNextCheck();
+
+    return () => {
+      if (intervalRef.current) {
+        clearInterval(intervalRef.current);
+      }
+      if (timeoutRef.current) {
+        clearTimeout(timeoutRef.current);
+      }
+    };
+  }, [enabled, isChecking, scheduleNextCheck]);
+
+  // Also handle CopilotKit runtime errors by listening to window events
+  useEffect(() => {
+    if (!enabled) return;
+
+    const handleCopilotKitError = (event: Event) => {
+      // Check if this is a CopilotKit-related error
+      const errorEvent = event as ErrorEvent;
+      if (
+        errorEvent.message?.includes('copilotkit') ||
+        errorEvent.message?.includes('CopilotKit') ||
+        errorEvent.filename?.includes('copilotkit')
+      ) {
+        markUnhealthy(`Runtime error: ${errorEvent.message}`);
+      }
+    };
+
+    window.addEventListener('error', handleCopilotKitError);
+    window.addEventListener('unhandledrejection', (event) => {
+      const reason = event.reason;
+      if (
+        typeof reason === 'string' && (
+          reason.includes('copilotkit') ||
+          reason.includes('CopilotKit') ||
+          reason.includes('ERR_CERT_COMMON_NAME_INVALID') ||
+          reason.includes('CORS')
+        )
+      ) {
+        markUnhealthy(`Unhandled promise rejection: ${reason}`);
+      }
+    });
+
+    return () => {
+      window.removeEventListener('error', handleCopilotKitError);
+    };
+  }, [enabled, markUnhealthy]);
+
+  return {
+    isHealthy,
+    isAvailable,
+    isChecking,
+    lastChecked,
+    errorMessage,
+    retryCount,
+    checkHealth,
+    markUnhealthy,
+  };
+};
+
diff --git a/frontend/src/hooks/usePhaseNavigation.ts b/frontend/src/hooks/usePhaseNavigation.ts
index 7e6e8805..6831ee2f 100644
--- a/frontend/src/hooks/usePhaseNavigation.ts
+++ b/frontend/src/hooks/usePhaseNavigation.ts
@@ -170,8 +170,8 @@ export const usePhaseNavigation = (
         // User is NOT in SEO phase - can progress to publish
         // This handles cases where user navigates away and comes back
         // Only auto-progress if user is already in a different phase (not actively in SEO)
-        if (currentPhase !== 'publish') {
-          setCurrentPhase('publish');
+      if (currentPhase !== 'publish') {
+        setCurrentPhase('publish');
         }
       }
     }
diff --git a/frontend/src/hooks/usePolling.ts b/frontend/src/hooks/usePolling.ts
index c45b5663..7c8c7162 100644
--- a/frontend/src/hooks/usePolling.ts
+++ b/frontend/src/hooks/usePolling.ts
@@ -153,7 +153,7 @@ export function usePolling(
         attemptsRef.current++;
       } catch (err) {
         const errorMessage = err instanceof Error ? err.message : 'Unknown error occurred';
-        console.error('Polling error:', errorMessage);
+        console.error('Polling error:', errorMessage, err);
         
         // Check if this is an axios error with subscription limit status
         // This is a fallback in case the interceptor doesn't catch it
@@ -161,15 +161,17 @@ export function usePolling(
         if (axiosError?.response?.status === 429 || axiosError?.response?.status === 402) {
           console.log('usePolling: Detected subscription error in axios error response', {
             status: axiosError.response.status,
-            data: axiosError.response.data
+            data: axiosError.response.data,
+            errorDataKeys: axiosError.response.data ? Object.keys(axiosError.response.data) : null
           });
           
           // Trigger subscription error handler (modal will show)
+          // Note: The interceptor may have already called this, but we call it again to be safe
           const handled = triggerSubscriptionError(axiosError);
           console.log('usePolling: triggerSubscriptionError returned', handled);
           
           if (handled) {
-            console.log('usePolling: Subscription error handled, stopping polling');
+            console.log('usePolling: Subscription error handled, stopping polling - modal should be visible');
             const errorMsg = axiosError.response?.data?.message || 
                            axiosError.response?.data?.error || 
                            'Subscription limit exceeded';
diff --git a/frontend/src/pages/ResearchTest.tsx b/frontend/src/pages/ResearchTest.tsx
new file mode 100644
index 00000000..13b14d73
--- /dev/null
+++ b/frontend/src/pages/ResearchTest.tsx
@@ -0,0 +1,239 @@
+import React, { useState } from 'react';
+import { ResearchWizard } from '../components/Research';
+import { BlogResearchResponse } from '../services/blogWriterApi';
+
+const samplePresets = [
+  {
+    name: 'AI Marketing Tools',
+    keywords: 'AI in marketing, automation tools, customer engagement',
+    industry: 'Technology',
+  },
+  {
+    name: 'Small Business SEO',
+    keywords: 'local SEO, small business, Google My Business',
+    industry: 'Marketing',
+  },
+  {
+    name: 'Content Strategy',
+    keywords: 'content planning, editorial calendar, content creation',
+    industry: 'Marketing',
+  },
+];
+
+export const ResearchTest: React.FC = () => {
+  const [results, setResults] = useState<BlogResearchResponse | null>(null);
+  const [showDebug, setShowDebug] = useState(false);
+  const [presetKeywords, setPresetKeywords] = useState<string[] | undefined>();
+  const [presetIndustry, setPresetIndustry] = useState<string | undefined>();
+
+  const handleComplete = (researchResults: BlogResearchResponse) => {
+    setResults(researchResults);
+  };
+
+  const handlePresetClick = (preset: typeof samplePresets[0]) => {
+    setPresetKeywords(preset.keywords.split(',').map(k => k.trim()));
+    setPresetIndustry(preset.industry);
+    setResults(null);
+  };
+
+  const handleReset = () => {
+    setPresetKeywords(undefined);
+    setPresetIndustry(undefined);
+    setResults(null);
+  };
+
+  return (
+    <div style={{ minHeight: '100vh', backgroundColor: '#f5f5f5' }}>
+      {/* Header */}
+      <div style={{
+        backgroundColor: '#1976d2',
+        color: 'white',
+        padding: '20px',
+        marginBottom: '20px',
+      }}>
+        <div style={{ maxWidth: '1400px', margin: '0 auto' }}>
+          <h1 style={{ margin: 0, fontSize: '28px' }}>🔬 Research Component Test Page</h1>
+          <p style={{ margin: '8px 0 0 0', fontSize: '14px', opacity: 0.9 }}>
+            Test the modular research wizard component
+          </p>
+        </div>
+      </div>
+
+      <div style={{ maxWidth: '1400px', margin: '0 auto', padding: '0 20px', display: 'flex', gap: '20px', flexWrap: 'wrap' }}>
+        {/* Left Panel - Controls */}
+        <div style={{ flex: '1 1 300px', minWidth: '300px' }}>
+          <div style={{
+            backgroundColor: 'white',
+            borderRadius: '8px',
+            padding: '20px',
+            marginBottom: '20px',
+            boxShadow: '0 2px 4px rgba(0,0,0,0.1)',
+          }}>
+            <h3 style={{ margin: '0 0 16px 0', color: '#333', fontSize: '18px' }}>
+              🎯 Quick Presets
+            </h3>
+            <div style={{ display: 'flex', flexDirection: 'column', gap: '8px' }}>
+              {samplePresets.map((preset, idx) => (
+                <button
+                  key={idx}
+                  onClick={() => handlePresetClick(preset)}
+                  style={{
+                    padding: '12px',
+                    backgroundColor: '#f0f7ff',
+                    border: '1px solid #b3d9ff',
+                    borderRadius: '6px',
+                    cursor: 'pointer',
+                    textAlign: 'left',
+                    fontSize: '14px',
+                    transition: 'all 0.2s ease',
+                  }}
+                  onMouseEnter={(e) => {
+                    e.currentTarget.style.backgroundColor = '#e3f2fd';
+                    e.currentTarget.style.borderColor = '#90caf9';
+                  }}
+                  onMouseLeave={(e) => {
+                    e.currentTarget.style.backgroundColor = '#f0f7ff';
+                    e.currentTarget.style.borderColor = '#b3d9ff';
+                  }}
+                >
+                  <div style={{ fontWeight: '600', color: '#1976d2', marginBottom: '4px' }}>
+                    {preset.name}
+                  </div>
+                  <div style={{ fontSize: '12px', color: '#666' }}>
+                    {preset.keywords}
+                  </div>
+                </button>
+              ))}
+            </div>
+            
+            <button
+              onClick={handleReset}
+              style={{
+                marginTop: '12px',
+                padding: '8px 16px',
+                backgroundColor: '#f5f5f5',
+                border: '1px solid #ddd',
+                borderRadius: '6px',
+                cursor: 'pointer',
+                fontSize: '13px',
+                width: '100%',
+              }}
+            >
+              ↻ Reset Test
+            </button>
+          </div>
+
+          {/* Debug Panel */}
+          <div style={{
+            backgroundColor: 'white',
+            borderRadius: '8px',
+            padding: '20px',
+            boxShadow: '0 2px 4px rgba(0,0,0,0.1)',
+          }}>
+            <div style={{ display: 'flex', justifyContent: 'space-between', alignItems: 'center', marginBottom: '12px' }}>
+              <h3 style={{ margin: 0, color: '#333', fontSize: '18px' }}>
+                🐛 Debug Panel
+              </h3>
+              <label style={{ cursor: 'pointer', fontSize: '14px' }}>
+                <input
+                  type="checkbox"
+                  checked={showDebug}
+                  onChange={(e) => setShowDebug(e.target.checked)}
+                  style={{ marginRight: '6px' }}
+                />
+                Show Debug
+              </label>
+            </div>
+
+            {showDebug && (
+              <div style={{
+                backgroundColor: '#f5f5f5',
+                borderRadius: '4px',
+                padding: '12px',
+                fontSize: '12px',
+                fontFamily: 'monospace',
+                maxHeight: '400px',
+                overflow: 'auto',
+              }}>
+                <pre style={{ margin: 0, whiteSpace: 'pre-wrap', wordBreak: 'break-word' }}>
+                  {JSON.stringify(results, null, 2)}
+                </pre>
+              </div>
+            )}
+          </div>
+        </div>
+
+        {/* Main Content - Wizard */}
+        <div style={{ flex: '2 1 800px' }}>
+          <ResearchWizard
+            initialKeywords={presetKeywords}
+            initialIndustry={presetIndustry}
+            onComplete={handleComplete}
+          />
+        </div>
+      </div>
+
+      {/* Footer Stats */}
+      {results && (
+        <div style={{
+          backgroundColor: 'white',
+          borderTop: '2px solid #e0e0e0',
+          padding: '20px',
+          marginTop: '40px',
+        }}>
+          <div style={{ maxWidth: '1400px', margin: '0 auto' }}>
+            <h3 style={{ margin: '0 0 16px 0', color: '#333', fontSize: '18px' }}>
+              📊 Research Statistics
+            </h3>
+            <div style={{ display: 'grid', gridTemplateColumns: 'repeat(auto-fit, minmax(200px, 1fr))', gap: '16px' }}>
+              <div style={{
+                backgroundColor: '#e3f2fd',
+                padding: '16px',
+                borderRadius: '8px',
+                border: '1px solid #90caf9',
+              }}>
+                <div style={{ fontSize: '12px', color: '#1976d2', fontWeight: '600', marginBottom: '4px' }}>
+                  Sources Found
+                </div>
+                <div style={{ fontSize: '28px', fontWeight: 'bold', color: '#1976d2' }}>
+                  {results.sources.length}
+                </div>
+              </div>
+
+              <div style={{
+                backgroundColor: '#f3e5f5',
+                padding: '16px',
+                borderRadius: '8px',
+                border: '1px solid #ce93d8',
+              }}>
+                <div style={{ fontSize: '12px', color: '#7b1fa2', fontWeight: '600', marginBottom: '4px' }}>
+                  Content Angles
+                </div>
+                <div style={{ fontSize: '28px', fontWeight: 'bold', color: '#7b1fa2' }}>
+                  {results.suggested_angles.length}
+                </div>
+              </div>
+
+              <div style={{
+                backgroundColor: '#e8f5e8',
+                padding: '16px',
+                borderRadius: '8px',
+                border: '1px solid #81c784',
+              }}>
+                <div style={{ fontSize: '12px', color: '#2e7d32', fontWeight: '600', marginBottom: '4px' }}>
+                  Search Queries
+                </div>
+                <div style={{ fontSize: '28px', fontWeight: 'bold', color: '#2e7d32' }}>
+                  {results.search_queries?.length || 0}
+                </div>
+              </div>
+            </div>
+          </div>
+        </div>
+      )}
+    </div>
+  );
+};
+
+export default ResearchTest;
+
diff --git a/frontend/src/services/blogWriterApi.ts b/frontend/src/services/blogWriterApi.ts
index 6d7860b3..455507ff 100644
--- a/frontend/src/services/blogWriterApi.ts
+++ b/frontend/src/services/blogWriterApi.ts
@@ -17,6 +17,23 @@ export interface ResearchSource {
   source_type?: string;
 }
 
+export type ResearchMode = 'basic' | 'comprehensive' | 'targeted';
+export type ResearchProvider = 'google' | 'exa';
+export type SourceType = 'web' | 'academic' | 'news' | 'industry' | 'expert';
+export type DateRange = 'last_week' | 'last_month' | 'last_3_months' | 'last_6_months' | 'last_year' | 'all_time';
+
+export interface ResearchConfig {
+  mode?: ResearchMode;
+  provider?: ResearchProvider;
+  date_range?: DateRange;
+  source_types?: SourceType[];
+  max_sources?: number;
+  include_statistics?: boolean;
+  include_expert_quotes?: boolean;
+  include_competitors?: boolean;
+  include_trends?: boolean;
+}
+
 export interface BlogResearchRequest {
   keywords: string[];
   topic?: string;
@@ -25,6 +42,8 @@ export interface BlogResearchRequest {
   tone?: string;
   word_count_target?: number;
   persona?: PersonaInfo;
+  research_mode?: ResearchMode;
+  config?: ResearchConfig;
 }
 
 export interface GroundingChunk {
diff --git a/frontend/src/services/blogWriterCache.ts b/frontend/src/services/blogWriterCache.ts
new file mode 100644
index 00000000..406ba4b9
--- /dev/null
+++ b/frontend/src/services/blogWriterCache.ts
@@ -0,0 +1,158 @@
+/**
+ * Blog Writer Cache Service
+ * 
+ * Provides persistent caching for outline and content to survive page refreshes
+ * and avoid unnecessary API calls. Shared by both CopilotKit and manual flows.
+ */
+
+interface CachedOutlineEntry {
+  outline: any[];
+  title_options?: string[];
+  research_keywords: string[];
+  created_at: string;
+}
+
+interface CachedContentEntry {
+  sections: Record<string, string>;
+  outline_ids: string[];
+  research_keywords: string[];
+  created_at: string;
+}
+
+class BlogWriterCacheService {
+  private readonly OUTLINE_CACHE_KEY = 'blog_outline';
+  private readonly TITLE_OPTIONS_CACHE_KEY = 'blog_title_options';
+  private readonly CONTENT_CACHE_PREFIX = 'blog_content_';
+
+  /**
+   * Get cached outline for research keywords
+   */
+  getCachedOutline(researchKeywords: string[]): { outline: any[]; title_options?: string[] } | null {
+    try {
+      if (typeof window === 'undefined') return null;
+
+      const savedOutline = localStorage.getItem(this.OUTLINE_CACHE_KEY);
+      const savedTitleOptions = localStorage.getItem(this.TITLE_OPTIONS_CACHE_KEY);
+
+      if (!savedOutline) {
+        return null;
+      }
+
+      const parsedOutline = JSON.parse(savedOutline);
+      if (!Array.isArray(parsedOutline) || parsedOutline.length === 0) {
+        return null;
+      }
+
+      // Basic validation: if we have an outline saved and it has sections, use it
+      // More sophisticated matching could compare research keywords if needed
+      const titleOptions = savedTitleOptions ? JSON.parse(savedTitleOptions) : undefined;
+
+      console.log(`Cache hit for outline (${parsedOutline.length} sections)`);
+      return {
+        outline: parsedOutline,
+        title_options: titleOptions
+      };
+    } catch (error) {
+      console.error('Error retrieving cached outline:', error);
+      return null;
+    }
+  }
+
+  /**
+   * Cache outline result
+   */
+  cacheOutline(outline: any[], titleOptions?: string[]): void {
+    try {
+      if (typeof window === 'undefined') return;
+
+      localStorage.setItem(this.OUTLINE_CACHE_KEY, JSON.stringify(outline));
+      if (titleOptions) {
+        localStorage.setItem(this.TITLE_OPTIONS_CACHE_KEY, JSON.stringify(titleOptions));
+      }
+      console.log(`Cached outline (${outline.length} sections)`);
+    } catch (error) {
+      console.error('Error caching outline:', error);
+    }
+  }
+
+  /**
+   * Generate cache key for content based on outline section IDs
+   */
+  private generateContentCacheKey(outlineIds: string[]): string {
+    const sortedIds = [...outlineIds].sort().join('|');
+    return `${this.CONTENT_CACHE_PREFIX}${sortedIds}`;
+  }
+
+  /**
+   * Get cached content for outline sections
+   */
+  getCachedContent(outlineIds: string[]): Record<string, string> | null {
+    try {
+      if (typeof window === 'undefined') return null;
+
+      const cacheKey = this.generateContentCacheKey(outlineIds);
+      const cachedContent = localStorage.getItem(cacheKey);
+
+      if (!cachedContent) {
+        return null;
+      }
+
+      const parsedSections = JSON.parse(cachedContent);
+      if (!parsedSections || typeof parsedSections !== 'object' || Object.keys(parsedSections).length === 0) {
+        return null;
+      }
+
+      // Verify that cached sections match outline structure
+      const cachedIds = new Set(Object.keys(parsedSections));
+      const outlineIdsSet = new Set(outlineIds.map(id => String(id)));
+      const idsMatch = outlineIdsSet.size === cachedIds.size &&
+                      Array.from(outlineIdsSet).every(id => cachedIds.has(id));
+
+      if (!idsMatch) {
+        console.log('Cached content does not match outline structure');
+        return null;
+      }
+
+      console.log(`Cache hit for content (${Object.keys(parsedSections).length} sections)`);
+      return parsedSections;
+    } catch (error) {
+      console.error('Error retrieving cached content:', error);
+      return null;
+    }
+  }
+
+  /**
+   * Cache content sections
+   */
+  cacheContent(sections: Record<string, string>, outlineIds: string[]): void {
+    try {
+      if (typeof window === 'undefined') return;
+      if (!sections || Object.keys(sections).length === 0) return;
+
+      const cacheKey = this.generateContentCacheKey(outlineIds);
+      localStorage.setItem(cacheKey, JSON.stringify(sections));
+      console.log(`Cached content (${Object.keys(sections).length} sections)`);
+    } catch (error) {
+      console.error('Error caching content:', error);
+    }
+  }
+
+  /**
+   * Check if content exists in state (helper for manual flow)
+   */
+  contentExistsInState(sections: Record<string, string>, outlineIds: string[]): boolean {
+    if (!sections || Object.keys(sections).length === 0) {
+      return false;
+    }
+
+    const existingIds = new Set(Object.keys(sections));
+    const outlineIdsSet = new Set(outlineIds.map(id => String(id)));
+    return outlineIdsSet.size === existingIds.size &&
+           Array.from(outlineIdsSet).every(id => existingIds.has(id));
+  }
+}
+
+// Export singleton instance
+export const blogWriterCache = new BlogWriterCacheService();
+export default blogWriterCache;
+
diff --git a/frontend/src/utils/navigationState.ts b/frontend/src/utils/navigationState.ts
new file mode 100644
index 00000000..2999a618
--- /dev/null
+++ b/frontend/src/utils/navigationState.ts
@@ -0,0 +1,161 @@
+/**
+ * Global Navigation State Utility
+ * 
+ * Manages navigation state preservation across subscription renewals and redirects.
+ * Supports:
+ * - Page path preservation
+ * - Phase state (for tools with phases like Blog Writer)
+ * - Tool-specific context (extensible for future tools)
+ */
+
+export interface NavigationState {
+  path: string;
+  phase?: string; // Phase ID for tools with phases (e.g., 'research', 'outline', 'content')
+  tool?: string; // Tool identifier (e.g., 'blog-writer', 'other-tool')
+  context?: Record<string, any>; // Tool-specific context data
+  timestamp: number; // When this state was saved
+}
+
+const NAVIGATION_STATE_KEY = 'subscription_navigation_state';
+
+/**
+ * Save navigation state before redirecting to pricing/subscription pages
+ * 
+ * @param path - Current page path (e.g., '/blog-writer')
+ * @param phase - Current phase ID (optional, for tools with phases)
+ * @param tool - Tool identifier (optional, defaults to detecting from path)
+ * @param context - Additional tool-specific context (optional)
+ */
+export const saveNavigationState = (
+  path: string,
+  phase?: string,
+  tool?: string,
+  context?: Record<string, any>
+): void => {
+  try {
+    // Auto-detect tool from path if not provided
+    const detectedTool = tool || detectToolFromPath(path);
+    
+    const state: NavigationState = {
+      path,
+      phase,
+      tool: detectedTool,
+      context,
+      timestamp: Date.now()
+    };
+    
+    sessionStorage.setItem(NAVIGATION_STATE_KEY, JSON.stringify(state));
+    console.log('[NavigationState] Saved navigation state:', state);
+  } catch (error) {
+    console.error('[NavigationState] Failed to save navigation state:', error);
+  }
+};
+
+/**
+ * Restore navigation state after returning from pricing/subscription pages
+ * 
+ * @returns NavigationState or null if not found/invalid
+ */
+export const restoreNavigationState = (): NavigationState | null => {
+  try {
+    const stored = sessionStorage.getItem(NAVIGATION_STATE_KEY);
+    if (!stored) {
+      return null;
+    }
+    
+    const state: NavigationState = JSON.parse(stored);
+    
+    // Validate state (must have path and reasonable timestamp)
+    if (!state.path || !state.timestamp) {
+      console.warn('[NavigationState] Invalid navigation state:', state);
+      return null;
+    }
+    
+    // Clear state after reading (one-time use)
+    sessionStorage.removeItem(NAVIGATION_STATE_KEY);
+    
+    console.log('[NavigationState] Restored navigation state:', state);
+    return state;
+  } catch (error) {
+    console.error('[NavigationState] Failed to restore navigation state:', error);
+    sessionStorage.removeItem(NAVIGATION_STATE_KEY);
+    return null;
+  }
+};
+
+/**
+ * Get navigation state without clearing it (for inspection)
+ */
+export const peekNavigationState = (): NavigationState | null => {
+  try {
+    const stored = sessionStorage.getItem(NAVIGATION_STATE_KEY);
+    if (!stored) {
+      return null;
+    }
+    
+    return JSON.parse(stored);
+  } catch (error) {
+    console.error('[NavigationState] Failed to peek navigation state:', error);
+    return null;
+  }
+};
+
+/**
+ * Clear navigation state (useful for cleanup)
+ */
+export const clearNavigationState = (): void => {
+  try {
+    sessionStorage.removeItem(NAVIGATION_STATE_KEY);
+    console.log('[NavigationState] Cleared navigation state');
+  } catch (error) {
+    console.error('[NavigationState] Failed to clear navigation state:', error);
+  }
+};
+
+/**
+ * Detect tool identifier from path
+ */
+const detectToolFromPath = (path: string): string | undefined => {
+  if (path.includes('/blog-writer') || path.includes('/blogwriter')) {
+    return 'blog-writer';
+  }
+  // Add more tool detection logic as needed
+  // if (path.includes('/other-tool')) {
+  //   return 'other-tool';
+  // }
+  return undefined;
+};
+
+/**
+ * Get current phase from localStorage for a specific tool
+ * This is a helper for tools that store phases in localStorage
+ */
+export const getCurrentPhaseForTool = (tool: string): string | null => {
+  try {
+    if (tool === 'blog-writer') {
+      return localStorage.getItem('blogwriter_current_phase') || null;
+    }
+    // Add more tool-specific phase retrieval as needed
+    return null;
+  } catch (error) {
+    console.error(`[NavigationState] Failed to get phase for tool ${tool}:`, error);
+    return null;
+  }
+};
+
+/**
+ * Save current phase to localStorage for a specific tool
+ * This is a helper for tools that store phases in localStorage
+ */
+export const saveCurrentPhaseForTool = (tool: string, phase: string): void => {
+  try {
+    if (tool === 'blog-writer') {
+      localStorage.setItem('blogwriter_current_phase', phase);
+      console.log(`[NavigationState] Saved phase '${phase}' for ${tool}`);
+    }
+    // Add more tool-specific phase saving as needed
+  } catch (error) {
+    console.error(`[NavigationState] Failed to save phase for tool ${tool}:`, error);
+  }
+};
+