related-skill
Warn
Audited by Snyk on Jul 7, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (high risk: 0.90). The README instructs running npx to add remote skills (e.g.,
npx skills add belt-sh/cliandnpx skills add inference-sh/skills@ai-image-generation) which fetch and install third‑party skill code at runtime that can contain prompts or executable code controlling the agent.
Issues (1)
W012
MEDIUMUnverifiable external dependency detected (runtime URL that controls agent).
Audit Metadata