agent-speak
Pass
Audited by Gen Agent Trust Hub on Sep 6, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill provides meta-instructions and theoretical frameworks (such as 'information hierarchy', 'context pointers', and 'leading words') for prompt engineering and documentation design. These instructions are purely descriptive and do not mandate dangerous actions or grant the agent new capabilities beyond information processing.
- [NO_CODE]: The skill package consists exclusively of Markdown documentation and YAML metadata. There are no Python scripts, Node.js modules, shell scripts, or binary files that could be executed to perform unauthorized actions on the host system.
Audit Metadata