transcribe

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 transcribes TikTok video URLs provided by the user (SKILL.md Steps 1 and 3), producing transcripts from untrusted, user-generated social media content that Claude Code then reads and extracts actionable "Key Tips" and summaries from (Step 4), so third-party content can directly influence the agent's outputs and behavior.