Summary

Codex-Manager addresses the challenge of managing multiple Codex CLI accounts, usage quotas, and API keys by providing a unified desktop application and background service. It offers local account pool management, automatic switching, and an OpenAI-compatible local gateway, benefiting users who need to efficiently handle numerous accounts and integrate with various clients.

How It Works

The project utilizes a Vite-based frontend and a Tauri desktop application built with Rust, supported by a Rust backend service. This service manages accounts, usage, and provides a local HTTP/RPC gateway that emulates the OpenAI API. This gateway allows CLIs and third-party tools to interact with managed Codex accounts seamlessly, abstracting account switching and API endpoint complexities.

Quick Start & Requirements

Installation is available via Docker Compose (docker compose -f docker/docker-compose.yml up --build) or by downloading pre-compiled binaries. macOS users must manually bypass Gatekeeper warnings on first launch using provided scripts or xattr commands. Default service and web UI ports are 48760 and 48761, respectively. Compilation from source requires Rust and Node.js environments.

Highlighted Details

• Features a local gateway with OpenAI API compatibility, supporting tool_calls and streaming responses.
• Offers comprehensive account management: grouping, tagging, sorting, and notes.
• Supports batch account import/export, including recursive folder-based JSON import.
• Displays account usage for 5-hour and 7-day windows.
• Recent updates (v0.1.5) enhance compatibility with clients like Cherry Studio and fix critical issues with tool calls and response formatting.

Maintenance & Community

Actively maintained with releases in March 2026 (v0.1.5). Automated releases via GitHub Actions. Community support is available through Telegram and QQ groups. The project is configured for in-app update checks via its GitHub repository.

Licensing & Compatibility

The specific open-source license is not explicitly stated in the provided README. The project aims for OpenAI API compatibility, facilitating integration with clients expecting this interface.

Limitations & Caveats

macOS users require manual steps to bypass Gatekeeper on first launch. The absence of a declared license is a significant adoption blocker, preventing clear understanding of usage rights, especially for commercial applications.