yt-seo
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill consists entirely of natural language instructions for the AI agent to follow when generating YouTube titles, descriptions, and tags. There are no scripts, executables, or command-line operations present in the skill.
- [DATA_EXPOSURE]: The skill references a specific file path (
~/.codex/youtube/voice.md) to read a user's voice profile. This is a functional requirement for the skill's stated purpose of writing in the user's style and does not target system credentials or sensitive environment variables. There are no network capabilities defined to exfiltrate this information. - [INDIRECT_PROMPT_INJECTION]: While the skill instructions involve processing external data (user videos for transcription), the skill lacks any dangerous capabilities—such as file writing, network access, or command execution—that could be exploited via malicious content in those videos.
Audit Metadata