self-assessment
Pass
Audited by Gen Agent Trust Hub on May 26, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill utilizes instruction-based prompt templates (
quick-assessment-prompt.txt,comprehensive-assessment-prompt.txt) to facilitate a knowledge assessment. These prompts contain no instructions to bypass safety filters, reveal system prompts, or override agent behavior. - [SAFE]: No network activity, data exfiltration patterns, or unauthorized file access was detected. The skill references local documentation paths (e.g.,
guide/ultimate-guide.md) and includes an optional feature to save results to a local file (assessment-results.json), which is consistent with its stated purpose. - [SAFE]: No obfuscation techniques (such as Base64 encoding, hex escapes, or hidden Unicode characters) were found. The skill's logic is transparently defined in the YAML frontmatter and associated text files.
- [SAFE]: The skill does not download external packages or execute remote code. It functions as a structured interaction within the agent's context.
Audit Metadata