Skill Vetter Bwm

Security-first skill vetting for AI agents. Use before installing any skill from ClawdHub, GitHub, or other sources. Checks for red flags, permission scope,...

This skill is a safe, coherent checklist you can use to manually vet other skills. Before relying on it: (1) verify the publisher/owner identity (the included _meta.json ownerId differs from the registry header and there's no homepage), (2) remember it only gives guidance — it won't automatically detect every issue, so perform an actual file-level code review when required, (3) when following its quick commands you will perform network queries (GitHub API) — ensure network access and rate limits are acceptable, and (4) do not allow the agent to automatically execute or install code it reviews; keep a human in the loop for high-risk decisions (credentials, elevated privileges, or obfuscated code). If you want higher assurance, clone target repos locally and inspect contents, check signatures, and cross-check author reputation externally.

Type: OpenClaw Skill Name: skill-vetter-bwm Version: 1.0.0 The skill is a defensive security tool designed to provide a structured vetting protocol for AI agents to evaluate other skills. It includes a comprehensive list of red flags, risk levels, and reporting templates to identify potential malware or vulnerabilities. The provided shell commands in SKILL.md are limited to fetching public repository metadata from the GitHub API for auditing purposes and do not exhibit any malicious behavior.