superseller-kie-discipline

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

• Third-party content exposure detected (high risk: 0.90). This skill's mandatory "Customer feedback capture protocol" instructs the agent to pull WhatsApp customer replies via WAHA GET /api//chats//messages, identify verbatim directives, append them to tenant rules, and then iterate—i.e., it ingests untrusted, user-generated content and uses it to drive subsequent actions, which can materially change behavior and therefore enables indirect prompt injection.