verification-before-completion
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOWNO_CODE
Full Analysis
- [NO_CODE] (SAFE): The skill consists entirely of Markdown instructions and does not contain any executable scripts, configuration files, or external dependency definitions.
- [PROMPT_INJECTION] (LOW): The content uses strong imperative language ('The Iron Law', 'Non-negotiable', 'lying') to override standard agent behavior regarding task reporting. While these terms are characteristic of behavioral steering and jailbreak-style language, their context here is strictly for quality assurance and enforcing task integrity rather than bypassing safety filters.
- [COMMAND_EXECUTION] (INFO): The skill defines a Gate Function requiring the identification and execution of verification commands. While the skill itself does not provide an execution engine, it establishes a protocol for an agent to interact with and rely upon command outputs.
Audit Metadata