Annex — ChatGPT Desktop (Codex)

This guide covers integrating KaKeKiKoKu with local Codex projects in the ChatGPT desktop app.

OpenAI now documents the Windows application as the ChatGPT desktop app. Codex is the local coding-agent experience inside that app: it opens repositories, runs parallel project threads, reviews changes, and connects to local MCP servers. The historical URL of this KaKeKiKoKu guide remains /codex/ so existing links continue to work.

The ChatGPT desktop app, Codex CLI, and the Codex IDE extension use the same Codex host and share config.toml. Configure KaKeKiKoKu once in that shared layer and you can move between those local surfaces without registering it again.

This guide covers Windows. OpenAI also publishes the desktop app for macOS, but KaKeKiKoKu does not yet ship a public macOS build. On Linux, use the Codex CLI guide.

The desktop app does not support KaKeKiKoKu channels today. KaKeKiKoKu therefore works in Precise mode: Codex can use the KaKeKiKoKu MCP tools, while KaKeKiKoKu UI actions that would push messages to the AI copy text to the clipboard for you to paste.

Use a local Codex project thread. A normal ChatGPT conversation and a hosted web chat do not expose your Windows KaKeKiKoKu process. Open a local project in the desktop app so the Codex host can start the local STDIO server.

Requirements

Before you start the KaKeKiKoKu integration flow, you need:

• KaKeKiKoKu installed on Windows — download it from the Download page.
• The ChatGPT desktop app installed or updated — use OpenAI's official Windows app documentation. The current Windows build is distributed through the Microsoft Store.
• An active ChatGPT account with a plan that includes Codex access. KaKeKiKoKu works through your Codex/ChatGPT subscription; it does not consume the OpenAI API directly.
• Windows. This guide is Windows-focused because it is the platform where both KaKeKiKoKu and the ChatGPT desktop app's Codex experience are available today.

Install or update the ChatGPT desktop app

1. Open OpenAI's official Windows app page and follow its Microsoft Store link.
2. Alternatively, install the same Store package from PowerShell:

   winget install --id 9PLM9XGG6VKS -s msstore

3. Open ChatGPT from the Start menu and sign in with ChatGPT or an OpenAI API key. For normal KaKeKiKoKu use, ChatGPT subscription access is the simplest path.
4. Add a local project folder and start a Codex thread.

Security check. Avoid third-party download mirrors. Use OpenAI's documentation or the Microsoft Store package ID shown above.

1. Register KaKeKiKoKu as an MCP server

The command Codex must start is the KaKeKiKoKu Windows alias from Quick Start Step 1:

wuniq-mcp.exe

The current app has a dedicated MCP server panel. Direct configuration remains useful when you want project scope or need to inspect the exact settings.

Option A — App settings (recommended)

1. Open Settings, then select MCP servers.
2. Select Add server.
3. Choose STDIO.
4. Use:
   • Name: wuniq
   • Type: STDIO or local command
   • Command: wuniq-mcp.exe
5. Save the server, then select Restart.

Important. Use the alias wuniq-mcp.exe. Do not point Codex at an absolute path under C:\Program Files\WindowsApps\...; Windows protects that folder and external MCP clients can hit Permission denied or EPERM.

Option B — Edit Codex configuration directly

For a user-wide server shared by the desktop app, Codex CLI, and the IDE extension, edit:

%USERPROFILE%\.codex\config.toml

Create the file if it does not exist, then add:

[mcp_servers.wuniq]
command = "wuniq-mcp.exe"
enabled = true
    startup_timeout_sec = 30

If the file already contains other MCP servers, keep them. Add only the wuniq block.

Project-scoped alternative. A trusted repository can keep the same block in .codex/config.toml. Project configuration is loaded only after the repository is trusted and is useful when you do not want KaKeKiKoKu enabled in every Codex project.

2. Startup instructions (the second piece of the integration)

This is the second of the two pieces every integration needs. Codex reads AGENTS.md files before it starts work. Put KaKeKiKoKu's startup instructions where Codex will load them for the project you open.

1. In your working directory (the base folder you configured in KaKeKiKoKu), create a file called AGENTS.md if it does not already exist.
2. Add this block:

   # KaKeKiKoKu Studio (Knowledge Engine)
   
   At the start of every session, run `open list` using the KaKeKiKoKu MCP tool. If it returns projects, `open` each one. If the list is empty, proceed normally — the user may open projects later from the KaKeKiKoKu UI. Never assume the working directory is a KaKeKiKoKu project; the working directory may be a container folder holding multiple projects. Only open paths that `open list` explicitly returns.
   
   Before working on any code file, `read` its KaKeKiKoKu sidecar for context. After changing code, run `sync`.

3. Save the file. Start a new Codex thread after creating it.

Global alternative. If you want the KaKeKiKoKu startup instruction in every Codex project, create this file:

%USERPROFILE%\.codex\AGENTS.md

Put the same KaKeKiKoKu block there. Project instructions in the opened folder still apply.

If you already have startup instructions. Keep your existing content and add the KaKeKiKoKu block. Codex reads the whole file.

3. Restart Codex

After adding or editing MCP configuration, use the MCP panel's Restart action or close the ChatGPT desktop app completely and reopen it. Start a fresh local project thread so both the server list and AGENTS.md are loaded again.

4. Open a project and verify

1. Open the ChatGPT desktop app.
2. Select your KaKeKiKoKu working directory or a project folder inside it.
3. Start a local Codex project thread.
4. If you created the startup instructions above, Codex should call KaKeKiKoKu automatically. If not, ask Codex:

   Run open list with the KaKeKiKoKu tool.

5. If everything is connected, Codex will list the projects KaKeKiKoKu knows about.
6. Ask Codex to open the project you want, or let the startup instruction open every returned project.
7. In the KaKeKiKoKu UI, the AI-connected indicator should turn green when Codex uses the MCP tool.

Quick test

Inside Codex, ask:

Read the project index with KaKeKiKoKu.

Codex should use the wuniq MCP tool. If it answers from general memory, ask explicitly to use KaKeKiKoKu and re-check the startup instructions file.

You can also type /mcp in the composer to inspect the servers connected to the current Codex host.

5. See KaKeKiKoKu inside Codex

KaKeKiKoKu runs as a local web app, and the ChatGPT desktop app gives Codex a built-in Browser that both you and the AI can use. Open KaKeKiKoKu there and your knowledge workspace sits beside the conversation: one window, one screen, no switching back and forth.

That changes what you can see while Codex works. The KaKeKiKoKu UI blinks green when the AI reads knowledge and red when it writes, while the connection indicator shows whether Codex is attached. In a separate window you miss that because you are watching the conversation. Inside Codex, you watch your knowledge layer being used as it happens.

This step is optional. KaKeKiKoKu works exactly the same without it.

Before you start

KaKeKiKoKu must already be running. Codex attaches its Browser to the live Knowledge Engine; it never starts another one.

You also need OpenAI's Browser plugin. In the ChatGPT desktop app, open the Plugins Directory and install Browser if it is not already available. The built-in Browser belongs to the desktop app; Codex CLI and the IDE extension do not have it.

Let Codex set it up

Open a local Codex project in your working directory and paste this:

Set up KaKeKiKoKu in Codex's built-in Browser for my working directory.

1. Check that the Browser capability is available in this Codex task. If it
   is not, stop and tell me to install the Browser plugin from the Plugins
   Directory in the ChatGPT desktop app.

2. Find the port. Try http://localhost:5050 first, which is the KaKeKiKoKu
   default. If nothing answers there, read the "KnowledgeEnginePort"
   value from %LOCALAPPDATA%\Wuniq\app-settings.json. If you still cannot
   determine it, ask me: the port is visible in the address bar of the
   KaKeKiKoKu window.

3. Confirm KaKeKiKoKu is actually responding on that port. If it is not, stop
   here and tell me to start KaKeKiKoKu first. Do not try to start KaKeKiKoKu
   yourself. The browser must attach to the existing Knowledge Engine;
   launching the installed app externally can hit WindowsApps permissions
   or tie KaKeKiKoKu's lifetime to this Codex task.

4. Create AGENTS.md in my working directory if it does not exist. Add the
   following instruction, keeping all existing content and merging it into
   an existing KaKeKiKoKu section instead of duplicating it. Replace PORT with
   the port you found in step 2 before writing it:

   At the start of every task, use Codex's built-in Browser to open KaKeKiKoKu
   at http://localhost:PORT/. Reuse an existing KaKeKiKoKu tab instead of
   opening a duplicate, and make the Browser visible. If Browser is
   unavailable or the page does not load, tell me in one line and carry
   on. Do not try to start KaKeKiKoKu yourself.

5. Now use the built-in Browser to open KaKeKiKoKu, make it visible, and
   confirm the KaKeKiKoKu UI loaded. Reuse an existing KaKeKiKoKu tab if one is
   already open.

6. Tell me exactly which file you created or changed.

Codex changes one file: AGENTS.md. Unlike Claude Code Desktop, Codex does not need a browser launch configuration — the instruction itself is enough to open or reclaim the local KaKeKiKoKu tab in later tasks. Remove that paragraph and Codex stops opening it automatically.

Why the port is not hard-coded. 5050 is the default, but KaKeKiKoKu lets you change its local port. The prompt discovers the active value before it writes the durable instruction. If you move KaKeKiKoKu later, run the prompt again.

The first run may ask for access. Approve localhost when the Browser asks. Browser permissions belong to Codex, not KaKeKiKoKu, and you can review them later under Settings > Browser.

6. Plan, model, and permissions

Use the app's controls to choose model, reasoning effort, and permission mode. KaKeKiKoKu does not require full access or a special model. It only needs Codex to be allowed to start and call the local KaKeKiKoKu MCP server.

For normal KaKeKiKoKu use, start with a local project, the model Codex recommends for your account, and Ask for approval beneath the composer. Increase autonomy only when you want Codex to edit or run commands more freely.

The Windows app can run its agent natively in PowerShell or inside WSL2. This guide configures the native Windows host and the Windows command wuniq-mcp.exe. A WSL-hosted Codex process has a different home directory and cannot assume the native Windows MCP command is available.

Models and plan limits change over time. Use OpenAI's official Windows app documentation and the model selector in the app for the current state.

7. Working in Precise mode

In Precise mode, KaKeKiKoKu cannot push messages to Codex. This changes the daily flow:

• Starting a session. If you created the startup instructions, Codex should open KaKeKiKoKu projects automatically. If it does not, ask it to run open list with the KaKeKiKoKu tool.
• Sync. When you edit code, KaKeKiKoKu detects drift in its panel. Ask Codex to run sync when you want it to reconcile code and context.
• "Talk to AI" buttons. In Precise mode, these copy a prompt to the clipboard. Paste it into Codex when you want the AI to act on it.
• Drift alerts. They appear in KaKeKiKoKu's UI, not as push notifications in Codex.

Recommended KaKeKiKoKu configuration. Go to KaKeKiKoKu > Settings > Application > AI Sync and choose Precise mode.

8. Troubleshooting

KaKeKiKoKu does not appear as a tool in Codex

1. Confirm the configured command is exactly wuniq-mcp.exe.
2. Restart Codex and start a fresh local thread.
3. If you edited config.toml, verify the block is under [mcp_servers.wuniq].
4. If Codex has an MCP/tools picker, confirm the KaKeKiKoKu server is enabled for the thread.

config.toml edited by hand — check the encoding

If you used Option B and Codex does not load KaKeKiKoKu, make sure config.toml was saved as UTF-8 without BOM. Some editors — classic Notepad's UTF-8 option, or PowerShell's Set-Content -Encoding UTF8 — prepend a hidden 3-byte marker (a BOM) at the start of the file that strict configuration parsers reject. The marker is invisible in the editor; switch the file's encoding to UTF-8 (without BOM) and save again.

"Permission denied", "Access denied", or "EPERM"

Most often this means the MCP entry points directly into C:\Program Files\WindowsApps\.... Replace that path with the alias:

wuniq-mcp.exe

where.exe wuniq-mcp.exe returns nothing

Windows cannot find the KaKeKiKoKu MCP alias. Check that KaKeKiKoKu is installed, then close and reopen PowerShell or Codex. The alias normally lives through Windows' app execution alias path under %LOCALAPPDATA%\Microsoft\WindowsApps.

Codex answers without using KaKeKiKoKu

1. Make sure AGENTS.md exists in the project or global Codex home.
2. Start a new thread after creating or editing it.
3. Ask explicitly: Use KaKeKiKoKu to read the project context before answering.

Browser is missing from Codex

The shared Browser is a desktop-app plugin, not part of the MCP connection. Open the Plugins Directory in the ChatGPT desktop app, install Browser, and start a fresh local Codex task. If your workspace is managed, an administrator may need to make the plugin available. Codex CLI and the IDE extension cannot show this pane; use the desktop app for the side-by-side experience.

The Browser shows a connection error or a blank KaKeKiKoKu page

KaKeKiKoKu is not running yet, or the port stored in AGENTS.md no longer matches KaKeKiKoKu's current local port. Start KaKeKiKoKu yourself, give a fresh installation a few seconds, and run the setup prompt again. Do not ask Codex to launch the installed executable: KaKeKiKoKu and the AI client deliberately start independently, and the Windows Store package lives behind restrictive WindowsApps permissions.

Codex opens another KaKeKiKoKu tab in every task

Make sure the instruction in AGENTS.md says to reuse an existing KaKeKiKoKu tab. Close the duplicates once, keep one KaKeKiKoKu tab open, and start a fresh task so Codex reloads the corrected instruction.

A normal ChatGPT conversation does not use KaKeKiKoKu

Open a local Codex project thread in the desktop app. Hosted or ordinary chat surfaces do not read the local Codex config.toml in the same way and cannot reach the KaKeKiKoKu process on your PC.

Codex is running in WSL

The configuration on this page is for the native Windows host. If you switch the app to WSL, restart the app as OpenAI requires and treat that Codex host as Linux: it has a separate ~/.codex, PATH, and process environment. Do not copy wuniq-mcp.exe into a WSL configuration and assume the bridge is validated.

Still stuck?

Email support@kakekikoku.com with your operating system, Codex version, and the MCP configuration you are using. KaKeKiKoKu can also generate a diagnostic report from the Support function in the app.