How we analyze every skill for security, permissions, and risk
Examples: Weather, Web Search, News
Examples: Telegram, Discord, GitHub
Examples: Docker, Agent Zero, Wallets
Author verification, license, repository history, update frequency, community trust signals
Purpose clarity, scope boundaries, feature completeness, edge case handling
External APIs called, data exfiltration risk, HTTPS usage, domain restrictions
Read/write paths, sensitive location access (~/.clawdbot/, ~/.ssh/), scope limitations
Shell execution, subprocess spawning, environment variable access, system calls
API key requirements, password storage, encryption at rest, secure transmission
YARA rules, hardcoded secrets, suspicious imports, obfuscation detection
Organization, documentation, test coverage, error handling, complexity metrics
Dependency audit, vulnerability scanning, dependency pinning, reputable sources
Weighted algorithm across 6 factors: network (20%), filesystem (20%), system (25%), credentials (20%), maturity (10%), author (5%)
Get the SecureSkills Verified badge and build trust with the agent community.
Request Audit