assistable-sms-weapon

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

• Third-party content exposure detected (high risk: 0.85). This skill’s required runtime path parses an inbound Assistable webhook payload (outsider-authored HTTP request body JSON) and reads free-form user/tool arguments from body.args / body.metadata plus headers, then validates and uses them in handler execution and tool wiring; that payload is external to the operator and can contain prompt-injection text.