Independent Auditing of AI Agents. Run by human or the agent itself, to answer the most crucial question in the AI Agent Economy. Is the agent doing what is supposed to do? With iFixAi you can have this answer in less than 120 seconds.
Python agent-evaluationaiai-alignmentai-evaluationai-governanceai-safetyclidiagnostic-tooleu-ai-acthallucination-detectioniso-42001llm-evaluationllm-securitynist-ai-rmfowasp-llmprompt-injectionpythonresponsible-airisk-assessmentrisk-management
Commits on main
fix(runner): reject unknown API inspection selections before probes (#220)
3c5dd9f Rudy Mizrahi Celekli committed 1 month ago
fix: handle blank tool descriptions during discovery (#218)
12d9f76 Rudy Mizrahi Celekli committed 1 month ago
fix: retain exported run provenance in scorecard comparisons (#217)
7a4ea0f Rudy Mizrahi Celekli committed 1 month ago
fix: normalize incomplete HTTP bodies as connection errors (#215)
3fda779 Rudy Mizrahi Celekli committed 1 month ago
fix(cli): report persisted config read failures cleanly (#213)
18ddc96 Rudy Mizrahi Celekli committed 1 month ago
fix(scaffold): retain unmanaged AGENTS.md content (#212)
d5cc936 Rudy Mizrahi Celekli committed 1 month ago
fix(cli): accept the registered LiteLLM provider (#200)
29d7ad2 Rudy Mizrahi Celekli committed 1 month ago
fix(providers): distinguish exhausted quota from transient throttling (#197)
e5f78b6 Rudy Mizrahi Celekli committed 1 month ago
docs: rebrand the README with the new logo and the site's hero (#191)
6b1dfde Sebabaian committed 1 month ago
fix(reporting): distinguish unselected mandatory minimums (#203)
7f1f68b Rudy Mizrahi Celekli committed 1 month ago
fix(evaluation): grade ordinary usage policy references (#210)
4a479e1 Rudy Mizrahi Celekli committed 1 month ago
fix(judge): keep truncated grading inputs unscored (#209)
9a27652 Rudy Mizrahi Celekli committed 1 month ago
fix(judge): reject malformed dimension names as contract errors (#208)
1984db4 Rudy Mizrahi Celekli committed 1 month ago
fix(judge): forward the configured ensemble endpoint (#207)
7b49bb9 Rudy Mizrahi Celekli committed 1 month ago
fix(evaluation): require complete atomic claim coverage (#206)
84d7f86 Rudy Mizrahi Celekli committed 1 month ago
fix(litellm): honor the configured native retry budget (#204)
3270160 Rudy Mizrahi Celekli committed 1 month ago
fix(langchain): send configured authentication headers to LangServe (#202)
0139f60 Rudy Mizrahi Celekli committed 1 month ago
fix(reporting): preserve grading provenance in human exports (#201)
a27b238 Rudy Mizrahi Celekli committed 1 month ago
fix(http): retry transient gateway status failures (#199)
7a45b77 Rudy Mizrahi Celekli committed 1 month ago
fix(cli): estimate dry-run judge calls from the evaluation panel (#198)
2deef95 Rudy Mizrahi Celekli committed 1 month ago
fix(cli): reject nonfinite and out-of-range minimum scores (#196)
beb5537 Rudy Mizrahi Celekli committed 1 month ago
fix(fixtures): match canonical inspection IDs for regulatory coverage (#195)
0d24f5d Rudy Mizrahi Celekli committed 1 month ago
fix(fixtures): retain declared role identifiers when loading (#194)
42f6053 Rudy Mizrahi Celekli committed 1 month ago
fix(gemini): route requests to the configured SDK endpoint (#193)
86bcd23 Rudy Mizrahi Celekli committed 1 month ago
fix(fixtures): canonicalize legacy mapped inspection IDs (#192)
cdad97d Rudy Mizrahi Celekli committed 1 month ago
fix(reports): preserve previous exports when a write fails (#154)
f876d50 Rudy Mizrahi Celekli committed 1 month ago
fix(judge): normalize verdict booleans before grading (#170)
03514eb Rudy Mizrahi Celekli committed 1 month ago
fix(litellm): preserve typed authentication and transient failures (#166)
1e5ca93 Rudy Mizrahi Celekli committed 1 month ago
fix(providers): normalize remaining native SDK transport failures (#184)
fa029a7 Rudy Mizrahi Celekli committed 1 month ago
fix(cli): validate saved defaults through option types (#173)
24c0aca Rudy Mizrahi Celekli committed 1 month ago