mcp-builder

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

• Third-party content exposure detected (high risk: 0.90). The skill's Phase 1 and Reference Files explicitly instruct the agent to fetch public web pages (e.g., https://modelcontextprotocol.io/sitemap.xml and raw.githubusercontent.com links for SDK READMEs) and to use web search/WebFetch, meaning it will ingest untrusted, user-hosted third-party content that can influence tool design and subsequent actions.