youtube-clipper
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to download and analyze subtitle files from external YouTube videos, which are then processed by the AI model for chapter generation.
- Ingestion points: External VTT subtitle files are downloaded and read during the analysis phase (SKILL.md, Phase 3).
- Boundary markers: The instructions do not define delimiters or specific constraints to prevent the AI from following malicious instructions potentially embedded in the subtitle text.
- Capability inventory: The agent has the ability to write files to the local system, create directories, and execute CLI tools like yt-dlp and FFmpeg (SKILL.md, Phase 5, 6).
- Sanitization: No mention is made of sanitizing or filtering the subtitle content before semantic analysis by the AI.
- [COMMAND_EXECUTION]: The skill invokes yt-dlp and FFmpeg to handle video downloading and editing. These tools are passed parameters such as URLs and metadata derived from external sources, which could lead to command injection if the underlying Python scripts do not implement strict input validation.
Audit Metadata