⚕️ the world's first unified, lifecycle-centric regulatory framework specifically designed for autonomous AI agents in medical environments (270+ requirements, 3 risk categories).

open-source machine-learning awesome framework ai fda artificial-intelligence cybersecurity ai-safety ai-agents medical-ai digital-health healthcare-ai regulatory-framework ai-regulation healthcare-compliance iso42001 fda-compliance aisvs qfire

• Updated Jun 11, 2026
• Python

Kayd — continuous threat modeling for agentic AI and classic systems. Scans your repo, maps trust boundaries, exports to OWASP Threat Dragon + SARIF.

maestro threat-modeling appsec stride threat-dragon ai-security agentic-ai aisvs

• Updated Jun 13, 2026

Blast radius modeling framework for agentic systems — 5 axes (action class, chain depth, external reach, reversibility window, identity scope), OSI cross-walk, and standards alignment.

osi ai-security nhi reversibility agentic agentic-ai blast-radius agent-identity agent-governance aisvs non-human-identity agentic-security decision-rights action-class

• Updated Jun 11, 2026

Reference implementation of OWASP AISVS C9.2.6 and C9.2.7: manifest-declared action class, deterministic gate, worst-case chain rule. JSON schema + Python.

python json-schema owasp cybersecurity reference-implementation ai-security reversibility agent-security aisvs action-class

• Updated Jun 9, 2026
• Python