using-systematic
Warn
Audited by Gen Agent Trust Hub on Mar 26, 2026
Risk Level: MEDIUMPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill uses aggressive, absolute language to override the agent's autonomous decision-making process.
- Evidence: Employs 'Override/bypass' markers such as 'EXTREMELY-IMPORTANT', 'ABSOLUTELY MUST', 'not negotiable', and 'not optional' to mandate specific tool usage.
- Evidence: Contains a 'Red Flags' section explicitly designed to suppress the agent's internal reasoning (e.g., 'you're rationalizing', 'Stop
- you're rationalizing') when determining task complexity or the need for clarification.
- Evidence: Commands the agent to ignore standard procedural steps, such as asking for context or performing codebase exploration, in favor of immediate skill tool invocation.
- Evidence: Establishes a '1% chance' rule that effectively bypasses the agent's default utility evaluation for tool execution.
Audit Metadata