mcp-builder

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

• Third-party content exposure detected (high risk: 0.85). The required workflow instructs using WebFetch to load public web content (e.g., https://modelcontextprotocol.io/llms-full.txt and GitHub README URLs) at runtime, which would be outsider-authored free text ingested into the agent’s LLM 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 "Use WebFetch to load: https://modelcontextprotocol.io/llms-full.txt", indicating the agent will fetch remote text at runtime that can directly control agent instructions/behavior.