task-brief-creator-caveman

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

• Third-party content exposure detected (medium risk: 0.30). The skill can ingest outsider-authored free text directly via the user’s chat or pasted planning notes during “When This Skill Runs,” then uses that content to guide Stage 1–3 codebase review and downstream brief construction (no filtering for “trusted first-party” beyond user-provided signal/anchors).