find-skills

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.75). The skill’s runtime workflow uses the Skills CLI to search/browse the open agent skills ecosystem (e.g., npx skills-il find [query] and installing from GitHub/other sources), which can ingest outsider-authored free text from public web/package listings into the agent’s context.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly instructs running runtime install commands like "npx skills-il add vercel-labs/agent-skills@vercel-react-best-practices", which will fetch and execute remote package code from the referenced repository (vercel-labs/agent-skills) at runtime.