MCP Integration Assistant

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

• Third-party content exposure detected (high risk: 1.00). The SKILL.md explicitly instructs building MCP integrations that fetch and expose external third‑party data (e.g., the Notion example's search-notion and get-notion-page tools that retrieve user-generated Notion pages/blocks via the Notion API) for the agent to read and act on, which meets the criteria for untrusted third-party content that can influence agent behavior.