HubLensAIYeachan-Heo/oh-my-codex
// archived 2026-04-21
56

// summary

oh-my-codex is a workflow enhancement layer designed to improve the functionality and consistency of the OpenAI Codex CLI. It provides specialized roles, reusable skills, and structured project state management to streamline complex development tasks. The tool is optimized for macOS and Linux environments, offering advanced features like team-based parallel execution and persistent completion loops.

// technical analysis

Oh-my-codex (OMX) serves as a specialized workflow layer designed to enhance the OpenAI Codex CLI experience by providing structured task management and persistent state tracking. It addresses the complexity of long-running development tasks by introducing standardized roles, skills, and a durable execution environment that keeps project plans and logs within a local .omx/ directory. The project prioritizes a consistent, agent-driven workflow, making a deliberate trade-off by focusing primarily on macOS and Linux environments to ensure reliable integration with tools like tmux.

// key highlights

01
Provides a standardized workflow layer using specific commands like $deep-interview and $ralplan to guide projects from clarification to completion.
02
Enables coordinated parallel execution through the $team command, allowing multiple agents to work on complex tasks simultaneously.
03
Maintains durable project state, including plans, logs, and memory, within a local .omx/ directory to ensure continuity across sessions.
04
Includes a built-in wiki feature that supports markdown-based, search-first documentation for local project context.
05
Offers robust diagnostic tools like omx doctor and omx exec to verify environment readiness and authentication before starting sessions.
06
Integrates seamlessly with native Codex hooks while providing an extensible plugin system via .omx/hooks/ for custom runtime behavior.

// use cases

01
Standardizing development workflows with deep-interview, planning, and execution skills
02
Coordinating parallel agent work using team-based runtime sessions
03
Managing project-specific guidance, logs, and memory through a structured .omx/ directory

// getting started

To begin, install the required packages globally using 'npm install -g @openai/codex oh-my-codex'. After installation, run 'omx setup' to configure the environment and 'omx doctor' to verify your system's readiness. Finally, launch your session with 'omx --madmax --high' to start the interactive workflow.