omc
Warn
Audited by Gen Agent Trust Hub on Apr 7, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONCREDENTIALS_UNSAFEREMOTE_CODE_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the user to add and install a plugin from a third-party GitHub repository (Yeachan-Heo/oh-my-claudecode).
- [EXTERNAL_DOWNLOADS]: The skill provides an alternative installation method via a global npm package named 'oh-my-claude-sisyphus', which differs from the project's primary name.
- [CREDENTIALS_UNSAFE]: The skill includes utilities for configuring external notification callbacks that require the user to provide sensitive information, such as Telegram bot tokens, chat IDs, and Discord webhooks.
- [COMMAND_EXECUTION]: The skill directs the agent to modify the experimental environment configuration for Claude Code by editing the '~/.claude/settings.json' file.
- [REMOTE_CODE_EXECUTION]: The core functionality involves complex multi-agent orchestration that executes parallel commands, automated verification loops, and persistent code fixes (team-exec, team-fix), creating a large surface for dynamic code execution.
- [EXTERNAL_DOWNLOADS]: The skill mentions optional dependencies from trusted organizations, specifically @google/gemini-cli and @openai/codex.
Audit Metadata