kg
Pass
Audited by Gen Agent Trust Hub on Jun 27, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill's primary function is to read and rank local markdown files based on topical relevance. Analysis of the source code (
scripts/kg.py) confirms that it relies solely on Python's standard library and has no capabilities for network communication or system modification. - [SAFE]: Input handling for the search topic is performed safely. The script tokenizes the input using a strict regular expression (
[a-z0-9_-]+), effectively sanitizing it before further processing. - [SAFE]: The skill's instructions for the AI agent are clear and technical, focusing on efficient data retrieval without attempting to bypass safety filters or override system prompts.
Audit Metadata