pm

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

• Third-party content exposure detected (medium risk: 0.30). The skill’s runtime loads board content (e.g., Notion pages, GitHub issues/projects items, Plane/Linear issues, or Obsidian card text) and processes it in modes like triage, next, status, and unblock, so outsider-authored free text from those systems is ingested without first selecting a specific item.