feature-models-roundtable

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

• Third-party content exposure detected (high risk: 0.70). Outsider free text can enter the LLM context via the user-provided feature request passed into models-roundtable (“Run models-roundtable on the feature request”), and that request is not authored by the operating user (it’s the operating user’s input, but the skill does not sanitize/limit it to user-authored internal sources).