mcp-security-hardening

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly fetches and processes tool descriptions and tool responses from external MCP servers (e.g., SecureMCPClient.list_tools calling self._client.list_tools(), MCPServerAnalyzer.analyze_server, and ToolIntegrityMonitor/validator checks of server tool descriptions), which are untrusted third-party inputs the agent is expected to read and interpret and therefore could enable indirect prompt injection.