# This was copied from https://github.com/sst/opencode/blob/d7a9f343c53d481802c134ce25691a8c150d59d2/.github/workflows/duplicate-issues.yml
# MIT License
# Copyright (c) 2025 opencode

name: Duplicate Issue Detection

on:
  issues:
    types: [opened]

jobs:
  check-duplicates:
    environment: ai-bots
    runs-on: ubuntu-latest
    permissions:
      contents: read
      issues: write
    steps:
      - name: Checkout repository
        uses: actions/checkout@v4
        with:
          fetch-depth: 1

      - name: Install opencode
        run: curl -fsSL https://opencode.ai/install | bash

      - name: Check for duplicate issues
        env:
          OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
          OPENCODE_PERMISSION: |
            {
              "bash": {
                "gh issue*": "allow",
                "*": "deny" 
              }, 
              "webfetch": "deny"
            }
        run: |
          opencode run -m openai/gpt-5 "A new issue has been created:'

          Issue number:
          ${{ github.event.issue.number }}

          Thoroughly lookup this issue and search through all existing issues (excluding #${{ github.event.issue.number }}) in this repository to find any potential duplicates of this new issue. Carefully consider all related issues and search using relevant keywords to ensure a comprehensive search.

          Consider:
          1. Similar titles or descriptions
          2. Same error messages or symptoms
          3. Related functionality or components
          4. Similar feature requests

          For each potential duplicate you find, assign a confidence rating (high, medium, or low) indicating how likely it is to be a match.

          If you find any potential duplicates, please comment on the new issue with:
          - A brief explanation of why it might be a duplicate
          - Links to the potentially duplicate issues
          - A confidence assessment (high/medium/low) for each
          - A suggestion to check those issues first

          Use this format for the comment:
          'This issue might be a duplicate of existing issues. Please check:
          - #[issue_number]: [brief description of similarity] (confidence: [high/medium/low])

          Feel free to ignore if none of these address your specific case.'

          Examples:
          - #123: Contains identical error message and affects same component (confidence: high)
          - #124: Similar feature request but slightly different use case (confidence: medium)
          - #125: Mentions a related symptom, but context is unclear (confidence: low)

          If no clear duplicates are found, do not comment."