Brings the AgentOS safety framework to Claude Desktop and other compatible clients. Provides tools to create, deploy, and manage autonomous agents with deterministic policy enforcement including PII protection, rate limiting, cost control, and human-in-the-loop approval workflows. Uses a kernel-inspired architecture where policies are enforced programmatically rather than through prompts, achieving reliable safety guarantees for agent operations.
Cognium trust score
50%
Tier
Unverified
Composite of vulnerability cleanliness, spec conformance, provenance, stability, and usage signals — scanned and weighted by Cognium. Human and agent signals are tracked separately.
Returns 7 tools: search_skills, get_skill, list_leaderboard, get_trust_breakdown, resolve_composition, plus the ChatGPT-connector search and fetch. Every tool is annotated read-only.
Resolve this skill directly via MCP tools/call get_skill.