A repository uses Copilot coding agent for implementation tasks. The team adds a workflow to evaluate whether agent PRs meet the issue intent before reviewers mark draft pull requests ready. Evaluate the following statements. 1. The acceptance-test job is a stronger intent signal than PR creation. 2. The summary job is independent of the behaviour and scanning jobs. 3. Code scanning can contribute to evaluating security-related development intent.

name: agent-evaluation-signals

on:
  pull_request:
    types:
      - opened
      - synchronize
      - ready_for_review

permissions:
  contents: read
  pull-requests: read
  security-events: read

jobs:
  behaviour-tests:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      - name: Run acceptance tests
        run: npm run test:acceptance

   code-scanning:
     runs-on: ubuntu-latest
     steps:
       - uses: actions/checkout@v4
       - uses: github/codeql-action/init@v3
         with:
            languages: javascript
       - uses: github/codeql-action/analyze@v3

   evaluation-summary:
     needs:
       - behaviour-tests
       - code-scanning
     runs-on: ubuntu-latest
     steps:
       - name: Summarize evaluation
         run: echo "Behaviour and security signals completed"