hook-analyzer

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

• Third-party content exposure detected (medium risk: 0.30). scripts/analyze_hook_segments.py directly ingests outsider-provided free text from a user-supplied breakdown.json (or stdin via cat ... | python ... -), reading fields like visual_content, speech_text, and frame_urls to construct the multimodal analysis context.