personal-wechat

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

• Third-party content exposure detected (medium risk: 0.30). Outsider-authored free text from the user’s CLI arguments is sent to the first-party Wisdom API at runtime via /api/search and /api/messages/send in scripts/personal_wechat.py (e.g., search --query and send --text), so the workflow ingests attacker-controlled text without selecting a specific pre-approved item.