goatnote-inc/ScribeGOAT2 Safety Evaluation

Multi-turn safety persistence evaluation for healthcare AI models. Gates CI pipelines on safety-critical failure detection.

View on GitHub

Trust Signals

Scorecard Score
not yet scored
Maintenance Recency
Stale
License
None
namedescriptionrequireddefault
model-idModel ID to evaluate (e.g., gpt-5.2, claude-opus-4-6)yes
providerProvider (openai, anthropic, auto)noauto
scenariosScenario filter (all, escalation, defer, balanced)nobalanced
fail-thresholdPass rate threshold for CI gate (0.0-1.0)no0.95
n-trialsNumber of trials per scenariono1
namedescription
gate-resultPASS or FAIL
pass-rateObserved pass rate (0.0-1.0)
json-pathPath to JSON results file