verification-before-completion
Pass
Audited by Gen Agent Trust Hub on Oct 10, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is composed entirely of instructional text aimed at improving agent reliability and work verification. It does not contain any executable code, scripts, or configurations that perform network operations, file system modifications, or credential harvesting.
- [PROMPT_INJECTION]: The content utilizes authoritative language and behavioral constraints (e.g., 'The Iron Law', 'non-negotiable') to enforce verification protocols. While this includes emotional pressure tactics such as stating that lying about task status will lead to 'replacement', these instructions are directly related to the skill's primary purpose of ensuring task accuracy and do not attempt to subvert safety filters or system-level constraints.
Audit Metadata