kimi-skill-creator
Pass
Audited by Gen Agent Trust Hub on Jun 19, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is a meta-utility for developing agent extensions. Its instructions focus on best practices, structured evaluation, and testing workflows.
- [COMMAND_EXECUTION]: Utilizes the
Bashtool for legitimate developer operations, including result aggregation with Python scripts and packaging skill directories using thezipcommand. - [DATA_EXFILTRATION]: No network-bound exfiltration patterns or sensitive file access was identified. File system interactions are limited to standard skill paths and local evaluation directories.
- [PROMPT_INJECTION]: The content does not attempt to subvert safety guidelines or override system prompts. All instructional language is directed toward the intended goal of skill creation.
Audit Metadata