api-tester

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

• Third-party content exposure detected (high risk: 0.80). The skill performs live requests to arbitrary API endpoints (see curl/fetch examples to https://api.example.com, the Postman collection with {{base_url}}, and code that calls response.json() and validates schemas), so the agent would ingest and interpret untrusted third‑party API responses as part of its workflow.