Home

Codexify

Let ChatGPT Web models, including 6 Astra Pro, work on the code on your own machine — safely, over an outbound-only tunnel, with no public URL.

Codexify is a local MCP bridge server. It runs on your computer, exposes a curated set of tools (read/write files, run shell commands, git, search, patching, long-running processes, and more), and connects those tools to ChatGPT through OpenAI's native Secure MCP Tunnel. ChatGPT can then act on your project the way Codex does — patch files in place, drive interactive processes, and keep a plan across a task. Structured filesystem tools stay pinned to the selected workspace; shell execution is deliberately unrestricted and has the service account's full authority.

This wiki is for people who run the server — no Rust knowledge required. It takes you from installation, through every command-line argument, to exactly what happens on each tool call.

---

Start here

If you want to…	Read
Get running in five minutes	Quick Start
Install the binary	Installation
Understand every flag and command	CLI Reference
Understand every config option	Configuration
Know what ChatGPT can actually do	Tools Reference
Understand what happens end-to-end	How It Works
Connect it to ChatGPT	Connecting to ChatGPT

Going deeper

• Multi-Project Mode — serve many repositories from one running server.
• Worktree Isolation — let two chats edit the same repo without colliding.
• Diff Checkpoints — show ChatGPT the full task diff or only what changed since the last diff.
• Context and Memory — keep a plan and notes across chats and restarts.
• AGENTS and Skills — teach ChatGPT your project's conventions and recurring tasks.
• Bridging MCP Servers — re-expose your other MCP servers through this one.
• Security Model — what a tool call can reach, and how to bound it.
• Audit Logging — record who called what, without leaking secrets.
• Troubleshooting and FAQ.

---

What you get

• Structured tools that behave the same on Windows, macOS, and Linux — read_file, write_file, glob, grep, tree, git operations — cheaper and safer than shelling out.
• Codex's own agent tools, faithfully ported — apply_patch, exec_command/write_stdin, view_image, update_plan, clock_*, skills_*.
• Codex's behavioural brief so the client edits carefully instead of rewriting files or running git reset --hard.
• Bounded output so a single tool call never floods ChatGPT's context.
• Persistent plan and notes that survive a new chat or a server restart.
• Exact-workspace continuation that can hand a dirty checkout, managed worktree, or projectless scratch workspace to a new ChatGPT conversation without cloning or resetting it.
• Provider-agnostic project cloning in multi-project mode for HTTPS/SSH .git URLs, with exact GitHub branch/PR/commit selection.
• MCP resource proxying so downloadable resources returned by bridged servers work through short-lived Codexify capabilities without exposing upstream URIs.
• Codex and Claude plugin skills folded into the same progressive skill catalogue.
• Verified self-update for standard installations, with detached service restart and rollback protection.
• Native outbound tunnel — the server listens only on 127.0.0.1, protects itself with a random per-process token, and never opens an inbound port.

One-line mental model

You point Codexify at a directory. Structured tools use that workspace boundary, while shell commands run unrestricted as the service account. Every call runs on your machine, and you can inspect, bound, and audit its activity.

Read Security Model before pointing it at anything sensitive.

---

Codexify is a Rust rewrite of the original Bun + TypeScript implementation, built on tokio + axum and the official rmcp SDK. It's open source under the MIT license.