omc-doctor
Warn
Audited by Gen Agent Trust Hub on Mar 24, 2026
Risk Level: MEDIUMCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill uses
node -eto execute inline Node.js scripts for file system operations, including searching for versions and deleting files. - [COMMAND_EXECUTION]: Provides automated fixes that execute destructive commands like
rm -rfon directories such as~/.claude/agents,~/.claude/commands, and~/.claude/skills, which may contain user-created content. - [EXTERNAL_DOWNLOADS]: Fetches markdown content from an external GitHub repository (
github.com/Yeachan-Heo/...) and uses it to overwrite the localCLAUDE.mdconfiguration file. - [COMMAND_EXECUTION]: Reads and modifies sensitive application configuration files (
~/.claude/settings.json) to identify and remove specific keys.
Audit Metadata