AgentForge Tier 2 Live #60
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| ## | |
| # AgentForge Tier 2 (live LLM) evals. | |
| # | |
| # Excluded from PR runs because each pass spends real OpenAI/Anthropic tokens. | |
| # Triggered nightly on the master branch and on demand via the Actions UI. | |
| # The runner refuses to start unless a live provider key is present, so a | |
| # fixture pass is never reported as live-provider proof. | |
| name: AgentForge Tier 2 Live | |
| on: | |
| workflow_dispatch: | |
| inputs: | |
| provider: | |
| description: 'Draft provider mode (openai or anthropic). Leave blank to auto-select from available secret.' | |
| required: false | |
| default: '' | |
| type: choice | |
| options: | |
| - '' | |
| - openai | |
| - anthropic | |
| schedule: | |
| # Nightly at 07:30 UTC (00:30 PT). Pinned to master. | |
| - cron: '30 7 * * *' | |
| concurrency: | |
| group: ${{ github.workflow }}-${{ github.ref }} | |
| cancel-in-progress: false | |
| jobs: | |
| tier2-live: | |
| name: Tier 2 - Live LLM evals | |
| runs-on: ubuntu-24.04 | |
| steps: | |
| - uses: actions/checkout@v6 | |
| - name: Setup PHP and Composer | |
| uses: ./.github/actions/setup-php-composer | |
| with: | |
| php-version: '8.2' | |
| - name: Run Tier 2 live evals | |
| env: | |
| AGENTFORGE_DRAFT_PROVIDER: ${{ github.event.inputs.provider }} | |
| AGENTFORGE_OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }} | |
| AGENTFORGE_ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }} | |
| run: php agent-forge/scripts/run-tier2-evals.php | |
| - name: Append Tier 2 summary to GitHub Actions | |
| if: ${{ !cancelled() }} | |
| run: bash agent-forge/scripts/ci/append-eval-step-summary.sh "Tier 2 — Live LLM evals" "agent-forge/eval-results/tier2-live-*.json" | |
| - name: Upload Tier 2 results | |
| if: ${{ !cancelled() }} | |
| uses: actions/upload-artifact@v7 | |
| with: | |
| name: agentforge-tier2-results | |
| path: agent-forge/eval-results/tier2-live-*.json |