yt-package
Pass
Audited by Gen Agent Trust Hub on Nov 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local Python script,
title.py, to perform linguistic and formatting analysis on YouTube metadata. This is a primary function of the skill and occurs locally. - [INDIRECT_PROMPT_INJECTION]: The skill processes content such as video titles and thumbnail text which are then used as arguments in shell commands for the
title.pyscript. - Ingestion points: User-provided or agent-generated YouTube titles and thumbnail text passed to
title.pyas arguments inSKILL.md. - Boundary markers: None are present in the command execution examples provided in the instructions.
- Capability inventory: The skill performs local script execution (
python3 title.py) and is instructed to read/write a voice profile file (~/.claude/youtube/voice.md) using agent tools. - Sanitization: No explicit escaping or sanitization of title/thumbnail strings is specified before they are used in the command line interface.
Audit Metadata