managed-deep-agents

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

• Potentially malicious external URL detected (high risk: 0.80). The skill explicitly calls and registers external MCP server URLs (e.g., "https://example.com/mcp") at runtime and shows a tool action that fetches external page content ("https://example.com") which can be returned into the agent's context and thus directly influence prompts/behavior.