pr-triage.yml

name: PR Triage

on:
  workflow_call:
    inputs:
      pr_number:
        description: "PR number to triage (optional, auto-detected from event)"
        required: false
        type: number
    secrets:
      COPILOT_GITHUB_TOKEN:
        required: true
        description: "Fine-grained PAT from a GitHub user with a Copilot license and Copilot Requests permission"
  # The following trigger is commented out since this is a reusable workflow.
  # Uncomment it if you want to use this workflow directly in this repository.
  # pull_request:
  #   types: [opened]

permissions:
  contents: read
  issues: write
  pull-requests: write

jobs:
  triage-pr:
    runs-on: ubuntu-latest
    timeout-minutes: 10
    permissions:
      contents: read
      issues: write
      pull-requests: write

    steps:
      - name: Set up Node.js
        uses: actions/setup-node@v4
        with:
          node-version: 22

      - name: Install Copilot CLI
        run: npm install -g @github/copilot

      - name: Collect PR context
        env:
          GH_TOKEN: ${{ github.token }}
          PR_NUMBER: ${{ inputs.pr_number || github.event.pull_request.number }}
        run: |
          set -euo pipefail

          gh label list \
            --repo "$GITHUB_REPOSITORY" \
            --limit 200 \
            --json name,description \
            > labels.json

          gh pr view "$PR_NUMBER" \
            --repo "$GITHUB_REPOSITORY" \
            --json number,title,body,files,labels,author,baseRefName,headRefName,isDraft,state,url \
            > pr.json

      - name: Select labels with Copilot
        env:
          COPILOT_GITHUB_TOKEN: ${{ secrets.COPILOT_GITHUB_TOKEN }}
        run: |
          set -euo pipefail

          copilot -s \
            --no-ask-user \
            --allow-tool=read \
            -p "$(cat <<'PROMPT'
          You are a GitHub pull request label triage assistant.

          Read ./pr.json and ./labels.json.

          Select labels that accurately describe the PR. Use only label names that already exist in labels.json.
          Do not invent labels. Do not comment on the PR. Do not ask questions.

          Prefer this taxonomy when the labels exist:
          - type: bug, type: feature, type: docs, type: build, type: ci, type: tests, type: maintenance, type: question
          - area:* labels for affected subsystem
          - platform:* labels only when the change is platform-specific
          - priority:* only when priority is explicit in the PR metadata
          - impact:* only for clear user-visible impact such as crash, memory leak, or visual diff
          - status:* only if already present or explicitly justified by the PR metadata

          Use the PR title, body, existing labels, and changed file paths.
          If unsure, choose fewer labels.

          Output only a JSON array of label names, for example:
          ["type: feature","area: rendering"]
          PROMPT
          )" > copilot-output.txt

          python3 <<'PY'
          import json
          import re
          from pathlib import Path

          raw = Path("copilot-output.txt").read_text()
          labels = {label["name"] for label in json.loads(Path("labels.json").read_text())}
          pr = json.loads(Path("pr.json").read_text())
          existing = {label["name"] for label in pr.get("labels", [])}

          try:
              parsed = json.loads(raw)
          except json.JSONDecodeError:
              match = re.search(r"```(?:json)?\s*(\[[\s\S]*?\])\s*```", raw)
              if match is None:
                  match = re.search(r"(\[[\s\S]*\])", raw)
              if match is None:
                  raise SystemExit(f"Copilot output did not contain a JSON array:\n{raw}")
              parsed = json.loads(match.group(1))

          if not isinstance(parsed, list):
              raise SystemExit(f"Copilot output must be a JSON array, got: {type(parsed).__name__}")

          selected = []
          for label in parsed:
              if not isinstance(label, str):
                  continue
              if label in labels and label not in existing and label not in selected:
                  selected.append(label)

          Path("selected-labels.json").write_text(json.dumps(selected, indent=2) + "\n")
          print(json.dumps(selected))
          PY

      - name: Apply labels
        env:
          GH_TOKEN: ${{ github.token }}
          PR_NUMBER: ${{ inputs.pr_number || github.event.pull_request.number }}
        run: |
          set -euo pipefail

          if [ "$(jq 'length' selected-labels.json)" -eq 0 ]; then
            echo "No new labels selected."
            exit 0
          fi

          labels="$(jq -c '{labels: .}' selected-labels.json)"
          gh api \
            --method POST \
            "/repos/$GITHUB_REPOSITORY/issues/$PR_NUMBER/labels" \
            --input - <<< "$labels"