remove-ai-marks

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

• Third-party content exposure detected (medium risk: 0.30). The required runtime workflow ingests outsider-authored free text directly from the skill user via the text pipeline (clipboard/file/stdin) and optionally sends it to a rewrite model through the scripts/rewrite_text.py prompt payload, so an attacker can submit poison text that the LLM consumes.