localsetup-skill-vetter

Vetting AI skills for security risks requires evaluating source provenance, codebase integrity, and permission scopes before installation. This skill enforces mandatory checks to detect red flags and risky patterns across public registries and GitHub repos.

An AI agent skill vetting protocol checks source provenance, performs code review, and assesses permission scopes. It generates a safety report with risk classification and clear remediation guidance to ensure safer skill onboarding.

Yes, you can use this vetting process on skills from public registries, GitHub repos, and inter-agent skill exchanges. It evaluates the source code and SKILL.md files to generate a comprehensive safety report.

To generate a safety report, run the vetting protocol on the skill's SKILL.md and its source code. The protocol evaluates source provenance, codebase integrity, and permission scopes before classifying the risk.

The best way to assess permission scopes is using a stepwise vetting protocol that evaluates source provenance and codebase integrity. It enforces mandatory checks to detect risky patterns and provides risk classification.

You need to review codebase integrity before adding an AI skill to prevent unsafe or unknown code from being installed. Vetting detects red flags and risky patterns, providing clear remediation guidance for safer skill onboarding.