skill-enable
Warn
Audited by Snyk on Jun 22, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (high risk: 0.80). The document instructs using the runtime command "gh skill install your-org/your-repo your-pkg-feature-a --agent claude-code", which fetches skill files from a remote GitHub repository (treated as a URL) that would be injected into the agent and thus directly control prompts.
Issues (1)
W012
MEDIUMUnverifiable external dependency detected (runtime URL that controls agent).
Audit Metadata