prompt-synthesis

Pass

Audited by Gen Agent Trust Hub on Aug 10, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is designed to be purely informational and instructional. It explicitly forbids the agent from performing any write, edit, delete, or shell operations. It instructs the agent to only use 'read-only tools' to ground its understanding of the workspace, which is a safe and standard practice. No malicious patterns, obfuscation, or data exfiltration vectors were identified.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 10, 2026, 09:29 AM
Security Audit — agent-trust-hub — prompt-synthesis