discordbot

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

• Third-party content exposure detected (medium risk: 0.30). The runtime workflow for this skill (the “send one opt-in DM” path) uses fixed, user-supplied IDs/keyword (recipient_id/consent_channel_id/consent_message_id/consent_keyword) and then reads the single corresponding Discord consent message via GET /channels/{consent_channel_id}/messages/{consent_message_id}, which is first-party Discord content but directly includes outsider-authored message text.