unjs

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

• Third-party content exposure detected (high risk: 0.80). The skill's generation guide and generation metadata explicitly state that package metadata, READMEs and documentation are fetched at runtime from public GitHub and npm APIs (see SKILL_GENERATION_GUIDE.md / GENERATION.md), which are untrusted, user-generated sources that the agent is expected to read and incorporate.