verification-before-completion

Pass

Audited by Gen Agent Trust Hub on Jun 18, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill consists exclusively of instructional text in English and Chinese. It does not include scripts, tools, or configuration that could execute code or access the file system.
  • [SAFE]: No prompt injection, data exfiltration, or obfuscation patterns were detected. The skill actually promotes security and accuracy by instructing the agent to verify all results and avoid making unverified claims.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 18, 2026, 11:31 AM
Security Audit — agent-trust-hub — verification-before-completion