remove-ai-marks

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

• Third-party content exposure detected (medium risk: 0.30). The skill’s required runtime workflow POSTs user-provided/pasted text (including clipboard content) into the service endpoints /inspect, /detect, and /clean as JSON ({"file":"<base64>","name":...}), so outsider-authored free text is directly ingested at runtime.