custom-sub-agents

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

• Third-party content exposure detected (high risk: 0.70). The skill's reference code and required guidance show agents that fetch and process arbitrary external APIs and services (e.g., DataProcessorAgent._fetch_from_api in references/01-agent-folder-structure.md and the integration test calling https://api.reporting-service.com/generate in references/04-testing-tools.md), and the SKILL explicitly lists "integration with external APIs/services", so the agent is expected to ingest untrusted third‑party content that can materially influence processing and follow-up actions.