mockup-lab

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

• Third-party content exposure detected (medium risk: 0.30). The skill’s required runtime workflow ingests outsider-authored free text via /interview/human-provided --feedback and variants --prompt, which are then passed into Stitch SDK/commands (e.g., ./run.sh iterate --feedback "<iterate_prompt>" and ./run.sh explore --views views.json), allowing an outsider to submit poison without selecting a specific item first.