kelly-mv

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

• Third-party content exposure detected (medium risk: 0.30). The required runtime workflow reads freeform “treatment”/“shot description”/“cast visual notes” text from the user-authored Busabase Bases (project/cast/shots) via busabase-provider.js and then uses it to build AI generation prompts in mv-model.js, so outsider authors can supply prompt-injection text by submitting records/fields in the workspace.