mcp-management

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly instructs browsing arbitrary web pages (e.g., the Gemini CLI example "Take a screenshot of https://example.com") and aggregates/discovers capabilities from multiple configured MCP servers into assets/tools.json that the LLM reads, which clearly involves ingesting untrusted public third‑party content.