claude-security-settings

Pass

Audited by Gen Agent Trust Hub on Apr 21, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides expert knowledge on configuring security permissions and shell protections within the Claude Code environment.
  • [SAFE]: Included examples of dangerous commands, such as command injection patterns or sensitive file access (e.g., rm -rf /, cat /etc/passwd), are explicitly labeled as blocked or prohibited behaviors.
  • [SAFE]: The setup instructions use standard local file operations like mkdir, cat, and echo to create configuration files (.claude/settings.json) within the project scope.
  • [SAFE]: No external network calls, remote dependencies, obfuscation techniques, or unauthorized privilege escalation attempts were detected.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 21, 2026, 01:17 AM