kutanti/LitmusAI Agent Evaluation

Run AI agent evaluations in CI/CD. Compare against baselines, detect regressions, post PR comments.

View on GitHub

Trust Signals

Scorecard Score
not yet scored
Maintenance Recency
Stale
License
None
namedescriptionrequireddefault
suiteTest suite name or path to YAML fileyes
agentAgent module path (e.g. my_agent:agent)yes
baselinePath to baseline results JSONno.litmus/baseline.json
thresholdMinimum pass rate to succeed (0.0-1.0)no0.7
budgetMaximum total cost budget ($)no""
concurrencyMax parallel evaluationsno5
save-baselineSave results as new baselinenofalse
runsNumber of evaluation runs for statistical analysisno1
log-dirDirectory to save result logsno""
post-commentPost results as PR commentnotrue
python-versionPython version to useno3.11
namedescription
pass-rateEvaluation pass rate (0.0-1.0)
total-costTotal evaluation cost ($)
passedNumber of passed tests
failedNumber of failed tests
has-regressionWhether regressions were detected
results-pathPath to results JSON file