found 1 skill in registry
Scans skills for tool poisoning, policy violations, and risk before registration or exposure. Detects hidden instructions, cross-tool shadowing, data exfiltration, and policy bypass patterns in SKILL.md, manifest, and schema files.