EdamAme-x

EdamAme-x / claudex

Public

use codex model on claude code

evex.land claude-code codex
18
2
69% credibility
Found Mar 06, 2026 at 18 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

claudex is a Bun-based launcher that runs Claude Code against an OpenAI-compatible endpoint, supporting local usage with Codex authentication files, ChatGPT tokens, and various environment variables for configuration.

Star Growth

See how this repo grew from 18 to 18 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is claudex?

ClaudeX is a TypeScript tool built on Bun that launches Anthropic's Claude Code—Claude AI's command-line coding agent—but proxies its API calls to OpenAI-compatible endpoints like ChatGPT's Codex backend. Developers drop in a ~/.codex/auth.json with tokens or API keys, tweak config.toml for providers, and run ./claudex to fire up Claude Code powered by models like gpt-5.3-codex, with auto token refresh and effort settings like xhigh reasoning. It solves the Claude vs ChatGPT dilemma for codex model change, letting Claude Code users tap OpenAI's code gen without rewriting tools.

Why is it gaining traction?

It stands out by auto-building multi-platform binaries (Linux, macOS, Windows) that track Claude Code's upstream releases daily, plus features like safe mode to block nonessential traffic, forced model overrides, and seamless ChatGPT bearer token handling. Devs love the drop-in proxy for codex github integration—no custom scripts needed for PR reviews or issues. The hook: Claude Oliver Rudolph-style tinkering with Claude KI on Codex endpoints, minus login hassles.

Who should use this?

CLI-heavy backend devs scripting with Claude Code who want cheaper Codex models over Anthropic's. Frontend teams doing codex github copilot swaps for boilerplate gen, or ops folks testing claude code in CI via codex github action. Perfect for codex github cli users eyeing pr review automation without vendor lock-in.

Verdict

Grab it if you're prototyping Claude Anthropic code flows on OpenAI infra—solid user-facing proxy and auth make it practical despite 18 stars signaling early maturity. 0.7% credibility score reflects niche appeal, but strong README, tests, and auto-releases suit tinkerers; watch for wider adoption.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.