Multi-Model Skill Trigger Eval #7
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
| name: Multi-Model Skill Trigger Eval | |
| on: | |
| schedule: | |
| # Run every Saturday at 08:00 UTC | |
| - cron: "0 8 * * 6" | |
| workflow_dispatch: | |
| inputs: | |
| models: | |
| description: "Comma-separated model aliases to evaluate (or 'all')" | |
| required: false | |
| default: "all" | |
| jobs: | |
| eval: | |
| name: Eval (${{ matrix.model }}) | |
| runs-on: ubuntu-latest | |
| permissions: | |
| contents: read | |
| strategy: | |
| fail-fast: false | |
| matrix: | |
| include: | |
| - model: sonnet | |
| concurrency: 20 | |
| - model: opus | |
| concurrency: 20 | |
| - model: gpt-4.1 | |
| concurrency: 5 | |
| - model: gemini-2.5-pro | |
| concurrency: 20 | |
| steps: | |
| - name: Check if model is selected | |
| id: check | |
| run: | | |
| MODELS="${{ github.event.inputs.models || 'all' }}" | |
| if [ "$MODELS" = "all" ] || echo ",$MODELS," | grep -q ",${{ matrix.model }},"; then | |
| echo "run=true" >> "$GITHUB_OUTPUT" | |
| else | |
| echo "run=false" >> "$GITHUB_OUTPUT" | |
| fi | |
| - name: Checkout code | |
| if: steps.check.outputs.run == 'true' | |
| uses: actions/checkout@v6 | |
| - name: Setup Deno | |
| if: steps.check.outputs.run == 'true' | |
| uses: denoland/setup-deno@v2 | |
| with: | |
| deno-version: v2.x | |
| - name: Setup Node.js | |
| if: steps.check.outputs.run == 'true' | |
| uses: actions/setup-node@v6 | |
| with: | |
| node-version: "24" | |
| - name: Run skill trigger evals | |
| if: steps.check.outputs.run == 'true' | |
| env: | |
| ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }} | |
| OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }} | |
| GOOGLE_API_KEY: ${{ secrets.GOOGLE_API_KEY }} | |
| run: deno run eval-skill-triggers --model ${{ matrix.model }} --concurrency ${{ matrix.concurrency }} |