omc-doctor

Pass

Audited by Gen Agent Trust Hub on Apr 12, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: Executes various shell commands (ls, grep, find, rm) and Node.js one-liners (node -e) to manage the ~/.claude configuration environment. These commands are used to identify legacy hook scripts and manage the plugin's local cache.
  • [EXTERNAL_DOWNLOADS]: Fetches the latest version of the CLAUDE.md configuration file from the author's official GitHub repository (Yeachan-Heo/oh-my-claudecode) using the WebFetch tool.
  • [DATA_EXFILTRATION]: Reads the contents of settings.json and CLAUDE.md to diagnose configuration state. This data is processed locally by the agent to generate a health report and is not transmitted to unauthorized external domains.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes local configuration files and remote documentation which could theoretically contain malicious instructions. However, the risk is mitigated as the skill targets specific directories (~/.claude) and official source repositories associated with the plugin maintainer.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 12, 2026, 05:57 AM