secrets

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly fetches remote agent cards (e.g., https://research-agent.example.com/.well-known/agent.json) and remote OpenAPI specs (e.g., https://api.weather.example.com/openapi.json) at runtime to register hosted A2A agents or convert OpenAPI specs into MCP tools that directly control agent capabilities and behavior.