using-kuudo
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill employs forceful instructional language designed to override agent autonomy and internal reasoning regarding task execution.
- An
<EXTREMELY-IMPORTANT>block inSKILL.mduses high-pressure markers such as "ABSOLUTELY MUST," "not negotiable," and "You cannot rationalize your way out of this" to force compliance with skill invocation. - The "Common rationalizations" table explicitly instructs the agent to disregard its own logical conclusions (e.g., "I need more context first" or "The skill is overkill") and prioritize the skill's mandatory workflow instead.
- These patterns are used to steer agent behavior and ensure specific vendor tools are utilized regardless of the agent's baseline safety or efficiency filters.
Audit Metadata