ci.yml

name: CI

on:
  push:
    branches: [main]
  pull_request:
    branches: [main]
  workflow_dispatch:
    inputs:
      run_evals:
        description: "Run eval harness (burns Anthropic tokens)"
        type: boolean
        default: false
      budget:
        description: "Budget in USD for eval run"
        type: string
        default: "5"

jobs:
  lint:
    name: Lint & type-check
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      - uses: actions/setup-python@v5
        with:
          python-version: "3.11"
      - name: Install dev dependencies
        run: |
          python -m pip install --upgrade pip
          pip install -e ".[dev]"
      - name: Ruff
        run: ruff check .
      - name: Mypy
        run: mypy agents models graph utils sandbox
        continue-on-error: true

  evals:
    name: Eval harness (manual)
    if: github.event_name == 'workflow_dispatch' && github.event.inputs.run_evals == 'true'
    runs-on: ubuntu-latest
    needs: lint
    steps:
      - uses: actions/checkout@v4
      - uses: actions/setup-python@v5
        with:
          python-version: "3.11"
      - name: Install dependencies
        run: |
          python -m pip install --upgrade pip
          pip install -e ".[dev]"
      - name: Build sandbox image
        run: docker build -t multi-agent-sandbox sandbox
      - name: Run evals
        env:
          ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
          EXPERIMENT_BUDGET_USD: ${{ github.event.inputs.budget }}
        run: python scripts/run_evals.py --out eval_results.json
      - name: Upload results
        if: always()
        uses: actions/upload-artifact@v4
        with:
          name: eval-results
          path: eval_results.json