youtube-to-heygen-video

Pass

Audited by Gen Agent Trust Hub on Apr 8, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns detected. The skill performs its stated function using well-known services and standard libraries. It does not attempt to exfiltrate sensitive data, gain persistence, or execute unauthorized code.
  • [INDIRECT_PROMPT_INJECTION]: The skill has an indirect prompt injection surface because it processes untrusted content from YouTube videos using an LLM.
  • Ingestion points: YouTube URLs are passed to the Gemini model for analysis in scripts/youtube_to_heygen_video.py.
  • Boundary markers: The prompt used for analysis (SCRIPT_EXTRACTION_PROMPT) lacks explicit delimiters or instructions to ignore embedded commands within the video content.
  • Capability inventory: The skill has network access (via requests) and database write access (via psycopg2).
  • Sanitization: The script performs basic string replacement for newline characters to facilitate natural speech synthesis but does not validate the semantic content of the generated script.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 8, 2026, 02:44 PM
Security Audit — agent-trust-hub — youtube-to-heygen-video