verification-before-completion
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill consists entirely of natural language instructions and a checklist designed to improve agent reliability by enforcing verification steps before marking tasks as complete.
- [NO_CODE]: There are no scripts, executables, or code blocks present in the skill, precluding common technical attack vectors like remote code execution or command injection.
- [REMOTE_CODE_EXECUTION]: No remote scripts or external code sources are referenced.
- [DATA_EXFILTRATION]: There are no network operations or instructions to read and send sensitive files.
- [PROMPT_INJECTION]: The instructions do not attempt to bypass safety guidelines or override system prompts; they align with standard operational best practices.
Audit Metadata