claude-optimizer

Pass

Audited by Gen Agent Trust Hub on Jul 1, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes common shell utilities including find, wc, and rg (ripgrep) to analyze CLAUDE.md files. These commands are used for their intended purpose of auditing documentation and do not involve untrusted input or elevated privileges.
  • [DATA_EXFILTRATION]: No network activity or external communication was detected. The skill operates entirely on local project documentation and does not attempt to access sensitive files like credentials, environment variables, or SSH keys.
  • [SAFE]: The instructions and workflow provide legitimate project management guidance. The skill contains no obfuscation, remote code execution, or attempts to override system safety protocols. Its behavior is consistent with its stated purpose of instruction optimization.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 1, 2026, 12:37 AM
Security Audit — agent-trust-hub — claude-optimizer