omc
Warn
Audited by Gen Agent Trust Hub on Nov 6, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSPERSISTENCECREDENTIALS_UNSAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill requires installing a plugin from a third-party GitHub repository (Yeachan-Heo/oh-my-claudecode) and suggests installing a global NPM package ('oh-my-claude-sisyphus'). These sources are not recognized as trusted organizations or well-known services.
- [EXTERNAL_DOWNLOADS]: The skill mentions optional dependencies from Google and OpenAI for extended CLI functionality. These are recognized as trusted organizations and well-known services.
- [PERSISTENCE]: The skill provides commands to enable an 'auto-resume daemon' (
omc wait --start), which indicates the creation of persistent background processes on the host system. - [CREDENTIALS_UNSAFE]: The documentation includes instructions for configuring Telegram and Discord notification callbacks, which require users to input sensitive credentials such as
<bot_token>and webhook URLs directly into the CLI tool. - [INDIRECT_PROMPT_INJECTION]: The skill possesses a significant attack surface for indirect injection as it processes untrusted user-defined tasks (e.g., 'autopilot: ') while maintaining high-privilege capabilities including Bash execution, file modification, and network access. The provided instructions do not include boundary markers or sanitization protocols to prevent the agent from executing malicious instructions embedded within the processed task data.
Audit Metadata