skill-forge

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

• Third-party content exposure detected (high risk: 0.80). SKILL.md and the examples explicitly call out using external research tools (e.g., the "Deep Research Agent" and Tavily via /sc:analyze) to fetch domain research such as "HIPAA regulations and healthcare compliance research" from public sources, and that fetched content is read and used to drive tool selection and design decisions — creating a clear pathway for untrusted third-party content to influence agent actions.