ideate

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

• Third-party content exposure detected (high risk: 0.70). The runtime workflow for /ideate includes “read AGENTS.md (the map skill writes it), README, and the top-reacted open issues” and also “read out/ideate/tracker.md if it exists”; top-reacted open issues/discussion text is typically outsider-authored, and that issue body/titles are read into the LLM context—an indirect prompt-injection surface.