claude-agent-sdk

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly configures and demonstrates calling external HTTP/SSE MCP servers and web fetches (e.g., SKILL.md "External MCP Servers (HTTP/SSE)" with url: "https://api.example.com/mcp" and templates/custom-mcp-server.ts fetching https://api.weather.com), so the agent ingests arbitrary third‑party web/API content that can influence its tool use and decisions.