cheez-read

Pass

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill implements a strict 'tilth-only' policy for file operations, effectively preventing the agent from using common shell tools like 'cat', 'ls', or 'find' which could be used for broader system exploration. It explicitly scopes operations to the current repository and lists system paths and user directories as out-of-scope. No evidence of prompt injection, data exfiltration, or obfuscation was found in the provided documentation or metadata.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 16, 2026, 03:14 AM
Security Audit — agent-trust-hub — cheez-read