verification-before-completion

Pass

Audited by Gen Agent Trust Hub on Mar 2, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: The skill defines best practices for agent reliability and honesty through verification steps. No malicious code or sensitive data access was found.
  • [NO_CODE]: The skill is comprised solely of markdown instructions and does not contain any executable scripts or binaries.
  • [PROMPT_INJECTION]: The use of strong directive language like 'The Iron Law' is intended to enforce task-level rigour and does not target the bypass of safety filters or extraction of system prompts.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 2, 2026, 04:11 AM