Runtime evidence layer for AI Act compliance. Intercepts agent tool calls, scores each with a conformal risk interval, writes a hash-chained audit record. Open source. Python.
python compliance attestation ai-safety ai-agents audit-trail conformal-prediction runtime-monitoring risk-scoring overt ai-governance llm-security eu-ai-act agent-security ai-act agent-governance article-14
-
Updated
May 19, 2026 - Python