actions-marketplace-validations/PruneTest
ML-backed CI test selection — run only the tests that matter for your diff
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Stale
- License
- None
Inputs
| name | description | required | default |
|---|---|---|---|
| api-key | Anthropic or OpenAI API key (for LLM layer) | yes | — |
| mode | shadow | conservative | standard | aggressive | no | shadow |
| test-dir | Path to test directory | no | __tests__ |
| config-file | Path to .testselector.yml | no | .testselector.yml |
| project-token | Project token from dashboard (enables history and ML layer) | no | — |
| test-results-file | Path to a JUnit/JSON test results file for historical learning | no | — |
| github-token | GitHub token for posting PR comments (requires pull-requests: write) | no | ${{ github.token }} |
Outputs
| name | description |
|---|---|
| blocking-tests | JSON array - run and block merge on failure |
| recommended-tests | JSON array - run but non-blocking |
| skip-tests | JSON array - skip this run |
| selection-confidence | Float 0-1: system confidence in this selection |
| reasoning | Plain text explanation of top decisions |
| estimated-time-saved | Estimated minutes saved vs running full suite |