openai-image-gen

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

• Third-party content exposure detected (medium risk: 0.30). Outsider-authored free text can be passed via the skill’s runtime CLI flag --prompt and is sent directly to the OpenAI Images API as the prompt field in scripts/gen.py:request_images() (then rendered into index.html via the prompt string), so the workflow ingests arbitrary user text at runtime without selecting a specific pre-defined item.