claude-code-course

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

• Third-party content exposure detected (high risk: 0.80). The repository explicitly shows agents and MCP servers that fetch public web content (e.g., skills/advanced/references/mcp-servers.md includes a weather_server that httpx.get()s https://wttr.in/{city} and the agent/tool docs list WebFetch/WebSearch), so the skill can retrieve and read untrusted, user-controlled third‑party content as part of its workflows.