runlit-dev/runlit eval

Evaluate AI-generated code for hallucinated APIs, intent drift, security issues, and compliance violations

View on GitHub

Trust Signals

Scorecard Score
not yet scored
Maintenance Recency
Stale
License
None
namedescriptionrequireddefault
api-keyrunlit API key (store in GitHub Secrets as RUNLIT_API_KEY)yes
block-thresholdScore below which the action fails the check (0-100)no50
warn-thresholdScore below which a warning annotation is posted (0-100)no70
api-urlrunlit API base URL (override for self-hosted)nohttps://api.runlit.dev
namedescription
scoreComposite eval score (0-100)
gradePASS, WARN, or BLOCK
eval-idUUIDv7 eval identifier for audit trail