mcp-integration

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

• Third-party content exposure detected (low risk: 0.10). Skill documentation describes MCP integrations (stdio/SSE/HTTP/WS) where the host connects to configured MCP servers and processes tool discovery/results, but it does not specify a runtime workflow that ingests outsider-authored free text without first selecting a specific item; the only explicit free-text ingestion path shown is user-provided command inputs.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.80). The skill contains explicit runtime MCP endpoints (e.g., https://mcp.asana.com/sse, https://api.example.com/mcp, wss://mcp.example.com/ws, https://mcp.github.com/sse) that the agent will connect to at runtime to discover tools and exchange messages, meaning external servers can directly influence agent behavior or provide executable tool responses.