content-amplifier

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

• Third-party content exposure detected (high risk: 0.75). The skill’s runtime workflow ingests outsider-authored free text when, in repurpose mode, the user provides “pasted transcript/caption/review text” from creator UGC (i.e., text authored by creators other than the operating user) and the agent reads it for atom extraction and repurposing.