data-collector

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

• Third-party content exposure detected (high risk: 1.00). The SKILL.md and CLI expose a --url option and explicitly "Fetches content from HTTP/HTTPS URLs" (example: https://jsonplaceholder.typicode.com/users), so the skill downloads and ingests arbitrary public third‑party web/API content supplied by users.