kernel-image-kadosabi
Pass
Audited by Gen Agent Trust Hub on Jul 13, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is a set of creative instructions for generating image prompts based on a specific fictional universe (Kadosabi). It contains no executable code, remote dependencies, or network operations.
- [DATA_EXPOSURE]: No sensitive file paths or credentials were found. The skill only uses the 'Read' tool to access its own internal reference documentation.
- [PROMPT_INJECTION]: The instructions are clear and focused on maintaining artistic style. There are no attempts to override safety filters or bypass agent constraints.
- [INDIRECT_PROMPT_INJECTION]: While the skill processes user-provided character names and descriptions to generate prompts, the attack surface is minimal. The agent is restricted to generating text prompts and reading local reference files, with no capabilities for file writing or network communication that could be exploited via injection.
Audit Metadata