mcp-to-skill

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

• Third-party content exposure detected (high risk: 0.90). The Phase 1 Discovery workflow explicitly instructs the agent to fetch MCP tool definitions from public package registries and repositories (e.g., "npm info " and "fetch the README or source") so the agent will read and interpret untrusted third‑party content that can directly influence classification and generated actions.