design

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

• Third-party content exposure detected (medium risk: 0.30). In the CIP Design sub-workflow, scripts/cip/generate.py and scripts/cip/core.py ingest outsider-authored free text at runtime (e.g., --brand, --industry, --style, --deliverable, --mockup) and incorporate it into LLM prompts for Gemini image generation via build_cip_prompt() → generate_with_nano_banana().