Prepare Validity Audit v0.4.0 (#18) #89
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: CI | |
| on: | |
| push: | |
| pull_request: | |
| permissions: | |
| contents: read | |
| jobs: | |
| test: | |
| runs-on: ubuntu-latest | |
| strategy: | |
| fail-fast: false | |
| matrix: | |
| python-version: ["3.11", "3.12"] | |
| steps: | |
| - uses: actions/checkout@v4 | |
| - uses: actions/setup-python@v5 | |
| with: | |
| python-version: ${{ matrix.python-version }} | |
| cache: pip | |
| - name: Install | |
| run: python -m pip install -e ".[dev,quant]" | |
| - name: Lint | |
| run: python -m ruff check validity_audit benchmarks tests protocol/ledger.py protocol/injected_bug_recall.py examples/self_contained/run_demo.py | |
| - name: Compile | |
| run: python -m compileall -q validity_audit protocol | |
| - name: Test | |
| run: python -m pytest -q | |
| - name: Console entry point | |
| run: validity-audit-ledger --ledger /tmp/validity-audit-ci-ledger.jsonl challenges | |
| - name: Legacy entry point | |
| run: python protocol/ledger.py challenges | |
| - name: Deterministic floor demonstration | |
| run: python benchmarks/injected/run.py | |
| - name: Self-contained scored golden case | |
| run: python golden_cases/self_contained/doc-bundle-01/run_case.py | |
| reproducibility: | |
| name: offline golden regression | |
| runs-on: ubuntu-latest | |
| steps: | |
| - uses: actions/checkout@v4 | |
| - uses: actions/setup-python@v5 | |
| with: | |
| python-version: "3.12" | |
| cache: pip | |
| - name: Install from clean checkout | |
| run: python -m pip install -e . | |
| - name: Run end to end with process network blocked and no API keys | |
| env: | |
| OPENAI_API_KEY: "" | |
| ANTHROPIC_API_KEY: "" | |
| GOOGLE_API_KEY: "" | |
| VALIDITY_AUDIT_REQUIRE_OFFLINE: "1" | |
| PYTHONPATH: benchmarks/offline_guard | |
| run: | | |
| python benchmarks/offline_guard/assert_blocked.py | |
| python golden_cases/self_contained/doc-bundle-01/run_case.py |