verification-before-completion

Pass

Audited by Gen Agent Trust Hub on Feb 26, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill defines a strict verification protocol requiring fresh evidence, such as test outputs and build status, before any success claims are made to the user.- [SAFE]: No prompt injection, data exfiltration, or obfuscation patterns were detected in the skill instructions or metadata.- [SAFE]: The content is explicitly designed to prevent shortcuts and rationalization, ensuring the agent remains honest and thorough in its operations.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 26, 2026, 10:12 PM