webmcp-user-interaction

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

• Third-party content exposure detected (high risk: 0.80). SKILL.md's "Before writing code" explicitly requires fetching live docs from https://webmachinelearning.github.io/webmcp/ and performing web searches (including community patterns and developer.chrome.com), so the agent is expected to ingest public third-party content that could materially influence tool implementation and runtime behavior.