mcp-builder

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

• Third-party content exposure detected (high risk: 0.90). The guide explicitly instructs the agent to fetch and read public web content (e.g., https://modelcontextprotocol.io/sitemap.xml and .md pages discovered from it, and raw.githubusercontent.com README.md files) as part of its workflow, which are open/public third‑party sources and can contain untrusted/user‑generated content that the agent will parse and interpret.

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

• Potentially malicious external URL detected (high risk: 0.90). The guide explicitly instructs the agent to WebFetch raw documentation at runtime (e.g. https://raw.githubusercontent.com/modelcontextprotocol/python-sdk/main/README.md and https://raw.githubusercontent.com/modelcontextprotocol/typescript-sdk/main/README.md), which would load remote markdown into the agent context and directly influence prompts/instructions and is treated as a required dependency—so it meets the runtime remote-content risk criteria.