adhd
Pass
Audited by Gen Agent Trust Hub on Aug 22, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill consists entirely of natural language instructions for output formatting and style adjustment intended to increase accessibility for readers with ADHD.
- [PROMPT_INJECTION]: The instructions include persistence markers ("These rules apply to every response for the rest of the session... If you are unsure whether they still apply, they do") to maintain the formatting style. These are benign persona-maintenance instructions. Critically, the skill includes an explicit safety clause stating that the system prompt and harness constraints outrank the skill's instructions.
- [SAFE]: No executable code, shell scripts, or external dependencies are included. While the markdown text contains examples of CLI commands (e.g.,
npm install jsonwebtoken) to illustrate formatting preferences, these are provided as sample output for a user and are not instructions for the agent to execute on the host system.
Audit Metadata