pezhik/skilltotal
Scan AI components (MCP servers, agent skills, npm/PyPI packages) for malware & risky capabilities. Local, deterministic, evidence-anchored. Free & OSS.
🎯 Vulnerability: Web Security Vulnerability
💡 Defense Fix: Implement strict input validation, contextual output encoding, and enforce least-privilege authorization controls.