verification-before-completion
Pass
Audited by Gen Agent Trust Hub on Apr 12, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No security issues detected. The skill consists entirely of markdown instructions designed to enforce a rigorous verification workflow for AI agents.
- [PROMPT_INJECTION]: The skill uses strong, directive language and behavioral constraints (e.g., "The Iron Law", "non-negotiable"). However, these are strictly used to improve the accuracy and reliability of the agent's work reporting rather than to bypass safety filters, extract system prompts, or induce prohibited behaviors. It serves as a process-oriented meta-instruction.
- [COMMAND_EXECUTION]: While the skill instructs the agent to run verification commands (such as tests or linters), it does not provide or execute any specific malicious commands itself. It refers to standard project tools expected to be present in the user's environment.
Audit Metadata