#3737th of 53 in Skills, plugins and rules
opencodex
Local proxy that lets Codex, Claude Code and Grok Build use any LLM
Documentation ↗
(opens in a new tab)Website ↗
(opens in a new tab)Repository on GitHub ↗
(opens in a new tab)
- Stars
- 17.4k
- License
- MIT
- Last commit
- Oct 2026
- Last release
- Oct 2026
- Language
- TypeScript
17 on 16 Jun 202617,396 on 13 Oct 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
opencodex is a local proxy that translates Codex's Responses API into the protocol a chosen provider speaks, covering streaming, tool calls, reasoning tokens and images. It also fronts Claude Code, Claude Desktop and Grok Build, with 40+ built-in providers or any OpenAI-compatible endpoint. A web dashboard on port 10100 configures providers, combos with failover, and a ChatGPT account pool with quota-aware routing.
Who it is for: Developers who want Codex or Claude Code to run on non-default LLM providers
Strengths
- 40+ built-in providers plus any OpenAI-compatible endpoint, including Ollama for local models
- Combos give one virtual model id with failover or weighted round-robin across providers
- Dashboard on localhost:10100 with live request log; `ocx stop` restores original Codex config
- Installs via npm with bundled Bun runtime; also desktop apps for macOS, Windows and Linux
Weaknesses
- ChatGPT account pooling may conflict with provider terms; README warns of possible enforcement
- Desktop app is beta and the Windows .msi is not code-signed
- Cursor provider is marked experimental
- JEV Auto routing is tested only against a mocked endpoint, not a live account
What it needs
- no GPU
- Docker + Compose
- Needs Node 18+, Bun (bundled on npm install)
- Models: Anthropic, Google Gemini, xAI Grok, OpenAI, DeepSeek
- port 10100
Also in Skills, plugins and rules
See all 53| Rank | Project | Score |
|---|---|---|
| 1 | skillsAgent skills for planning, TDD, debugging and code review | 94 out of 100 |
| 2 | eccAgents, skills, hooks and rules bundle for Claude Code, Codex and other harnesses | 93 out of 100 |
| 3 | cavemanCuts token use in coding agents with terse replies and compressed inputs | 93 out of 100 |
| #4 | ponytailSkill prompt that makes coding agents write less code and add tests | 91 out of 100 |
| #5 | ui-ux-pro-max-skillDesign-system skill that gives coding agents searchable UI styles, palettes and fonts | 90 out of 100 |