watch

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly fetches and ingests untrusted, user-generated YouTube content (metadata via scripts/get-metadata.ts and captions via scripts/get-cc.ts) and then sends that content and the video URL to Gemini as part of the transcribe workflow (scripts/transcribe.ts and SKILL.md), allowing third-party content to influence subsequent analysis, indexing, and actions.