llmx/.github/workflows/issue-deduplicator.yml

name: Issue Deduplicator

on:
  issues:
    types:
      - opened
      - labeled

jobs:
  gather-duplicates:
    name: Identify potential duplicates
    if: ${{ github.event.action == 'opened' || (github.event.action == 'labeled' && github.event.label.name == 'codex-deduplicate') }}
    runs-on: ubuntu-latest
    permissions:
      contents: read
    outputs:
      codex_output: ${{ steps.codex.outputs.final-message }}
    steps:
      - uses: actions/checkout@v5

      - name: Prepare Codex inputs
        env:
          GH_TOKEN: ${{ github.token }}
        run: |
          set -eo pipefail

          CURRENT_ISSUE_FILE=codex-current-issue.json
          EXISTING_ISSUES_FILE=codex-existing-issues.json

          gh issue list --repo "${{ github.repository }}" \
            --json number,title,body,createdAt \
            --limit 1000 \
            --state all \
            --search "sort:created-desc" \
            | jq '.' \
            > "$EXISTING_ISSUES_FILE"

          gh issue view "${{ github.event.issue.number }}" \
            --repo "${{ github.repository }}" \
            --json number,title,body \
            | jq '.' \
            > "$CURRENT_ISSUE_FILE"

      - id: codex
        uses: openai/codex-action@main
        with:
          openai-api-key: ${{ secrets.CODEX_OPENAI_API_KEY }}
          allow-users: "*"
          model: gpt-5
          prompt: |
            You are an assistant that triages new GitHub issues by identifying potential duplicates.

            You will receive the following JSON files located in the current working directory:
            - `codex-current-issue.json`: JSON object describing the newly created issue (fields: number, title, body).
            - `codex-existing-issues.json`: JSON array of recent issues (each element includes number, title, body, createdAt).

            Instructions:
            - Compare the current issue against the existing issues to find up to five that appear to describe the same underlying problem or request.
            - Focus on the underlying intent and context of each issue—such as reported symptoms, feature requests, reproduction steps, or error messages—rather than relying solely on string similarity or synthetic metrics.
            - After your analysis, validate your results in 1-2 lines explaining your decision to return the selected matches.
            - When unsure, prefer returning fewer matches.
            - Include at most five numbers.

          output-schema: |
            {
              "type": "object",
              "properties": {
                "issues": {
                  "type": "array",
                  "items": {
                    "type": "string"
                  }
                },
                "reason": { "type": "string" }
              },
              "required": ["issues", "reason"],
              "additionalProperties": false
            }

  comment-on-issue:
    name: Comment with potential duplicates
    needs: gather-duplicates
    if: ${{ needs.gather-duplicates.result != 'skipped' }}
    runs-on: ubuntu-latest
    permissions:
      contents: read
      issues: write
    steps:
      - name: Comment on issue
        uses: actions/github-script@v8
        env:
          CODEX_OUTPUT: ${{ needs.gather-duplicates.outputs.codex_output }}
        with:
          github-token: ${{ github.token }}
          script: |
            const raw = process.env.CODEX_OUTPUT ?? '';
            let parsed;
            try {
              parsed = JSON.parse(raw);
            } catch (error) {
              core.info(`Codex output was not valid JSON. Raw output: ${raw}`);
              core.info(`Parse error: ${error.message}`);
              return;
            }

            const issues = Array.isArray(parsed?.issues) ? parsed.issues : [];
            const currentIssueNumber = String(context.payload.issue.number);

            console.log(`Current issue number: ${currentIssueNumber}`);
            console.log(issues);

            const filteredIssues = issues.filter((value) => String(value) !== currentIssueNumber);

            if (filteredIssues.length === 0) {
              core.info('Codex reported no potential duplicates.');
              return;
            }

            const lines = [
              'Potential duplicates detected. Please review them and close your issue if it is a duplicate.',
              '',
              ...filteredIssues.map((value) => `- #${String(value)}`),
              '',
              '*Powered by [Codex Action](https://github.com/openai/codex-action)*'];

            await github.rest.issues.createComment({
              owner: context.repo.owner,
              repo: context.repo.repo,
              issue_number: context.payload.issue.number,
              body: lines.join("\n"),
            });

      - name: Remove codex-deduplicate label
        if: ${{ always() && github.event.action == 'labeled' && github.event.label.name == 'codex-deduplicate' }}
        env:
          GH_TOKEN: ${{ github.token }}
          GH_REPO: ${{ github.repository }}
        run: |
          gh issue edit "${{ github.event.issue.number }}" --remove-label codex-deduplicate || true
          echo "Attempted to remove label: codex-deduplicate"