skill-vetter

To vet third-party AI skills before installation, you need a formal protocol that assesses source credibility, performs code review for red flags, evaluates permission scope, and outputs a structured risk classification report with a final safety verdict.

Permission scope evaluation in AI skill risk assessment is the process of analyzing a skill's file read/write access, network requirements, and command execution privileges to ensure it maintains minimal necessary permissions and mitigates potential system threats.

Checking AI skills for red flags requires a mandatory code review of all files from sources like GitHub or ClawdHub to identify malicious patterns, evaluate author reputation, and classify potential data handling vulnerabilities before installation.

Yes, you can apply an automated vetting process to AI skills obtained from any source. The protocol assesses origin trustworthiness, recency, permission scopes, and code-level risks to produce a structured report with a clear installation verdict.

The best way to classify risks in third-party AI skills is to enforce a formal vetting process that evaluates source credibility, executes code review, and assesses data handling scenarios to generate a structured report with a defined risk level and safety notes.