I Found the Best Claude Code GUI (That Works with Codex & Gemini Too)

If you want to keep coding and forget the AI coding agent wars, you need to try one of the best AI agent GUIs out there.

Luckily, I found one and tested it, so you don’t have to waste your time.

It started as a lazy afternoon, feeling a little bored but still keeping up with my tests and reviews. Today I’ve been working on a project that requires both Claude Code and Codex.

Switching between terminals, managing sessions, and keeping track of what each agent is doing across projects.

I know there are several GUI apps for Claude Code, but when I found this one, I immediately fell in love

CloudCLI (aka Claude Code UI) is one of those projects that has earned its reputation on GitHub with 11,000+ stars and is still gaining traction for the right reasons.

What is CloudCLI?

CloudCLI is a free, open-source web UI that gives you a full graphical interface for Claude Code, OpenAI Codex, Cursor CLI, and Gemini CLI.

You get one clean dashboard where you can:

• Switch between AI agents from a single interface
• Manage projects and sessions visually
• Browse files, run shell commands, and handle git
• Access your coding agents from any device, including mobile

The setup is one command:

npx @cloudcli-ai/cloudcli

It auto-discovers your existing Claude Code sessions from your ~/.claude folder, so you're not starting from scratch.

Most GUI tools I’ve tested focus on Claude Code, CloudCLI connects to Claude Code, Codex, Cursor CLI, and Gemini CLI.

For someone like me who regularly tests and switches between agents, this is what I needed.

In this article, I’ll walk you through the setup, show you every tab and feature I tested, and share my honest take on whether CloudCLI deserves the hype.

Let’s start with the installation.

Setting Up CloudCLI

You need Node.js v22+ installed on your machine.

Open your terminal and run:

npx @cloudcli-ai/cloudcli

It installs the package, applies the database schema, configures web push notifications, and starts the server.

You’ll see two URLs in your terminal:

• Production mode: [http://localhost:3001](http://localhost:3001)
• Development mode: [http://localhost:5173](http://localhost:5173)

For regular use, you can also install it globally:

npm install -g @cloudcli-ai/cloudcli
cloudcli

Once the server is running, open http://localhost:3001 in your browser.

Git Configuration

The first screen asks you to configure your Git identity, your name and email.

These are saved as git config --global user.name and git config --global user.email.

This makes sense since CloudCLI has built-in source control and needs your Git identity for commits.

Fill in your details and hit Next.

Connecting Your AI Agents

The next screen shows you all the AI agents you can connect.

Each one has a Login button or shows a green checkmark if already authenticated.

In my setup, I connected:

• Claude Code — Requires OAuth login with your Claude account
• Cursor — Needs the Cursor CLI installed
• OpenAI Codex — Connected with my account
• Gemini — Connected with my Google account

You can connect one or all four, and you can always configure them later in Settings. I hit “Complete Setup” and landed on the main interface.

First Look at the Dashboard

The main interface is clean and well-organized.

On the left sidebar, you get:

• Projects — Lists all your local projects
• Conversations — Your chat history with each agent
• Search — Quick project search with Ctrl+K

The center area is your chat workspace. You select your AI provider and model, then start a conversation.

Model Selection

Click on the model name to open the model selector. This is where the multi-agent support shines.

Under Anthropic, I can see:

• Opus, Sonnet, Haiku
• Opus 4.6, Opus Plan
• Sonnet [1M], Opus [1M]

Under OpenAI:

• GPT-5.5, GPT-5.4

You pick the model and type your prompt

Interface — Chat, Files, Shell & Source Control

CloudCLI has four major tabs across the top: Chat, Shell, Files, and Source Control.

Chat

The chat tab is your main workspace. It works like Claude Code’s terminal interface but with a visual layer on top.

You type prompts, see responses with syntax highlighting, and manage your conversation history in the sidebar.

At the bottom, you’ll notice:

• Default Mode toggle — switch between modes
• Context usage — shows your token consumption
• A message counter badge

Type / to access slash commands. CloudCLI supports the same built-in commands you'd use in Claude Code's terminal.

Some commands include:

• /help — Show help documentation for Claude Code
• /clear — Clear the conversation history
• /model — Switch or view the current AI model
• /cost — Display token usage and cost information
• /memory — Open CLAUDE.md memory file for editing

These are the same commands from Claude Code’s CLI.

File Explorer

Click the Files tab, and you get a full file browser for your project.

It shows your project’s directory structure with file names, sizes, modification dates, and permissions.

You can browse, search, and edit files with syntax highlighting

Shell — Running Claude Code Inside the GUI

This is my favourite feature.

Click the Shell tab, and you get a full terminal embedded in the browser. Claude Code runs directly inside CloudCLI.

In my test, it launched Claude Code v2.1.146 with Sonnet 4.6, connected to my organization account.

You can run Claude Code from your phone or tablet through the browser without SSH-ing into your machine.

For remote development workflows, start a session on your desktop and continue it from your phone.

Source Control

The Source Control tab gives you a built-in Git interface.

You get three sub-tabs:

• Changes — Stage and commit files with a commit message
• Commits — View commit history
• Branches — Switch between branches

I clicked on the branch dropdown and saw all my branches. You can switch branches, create new ones, fetch remote changes, and commit

Settings, Customization & Plugins

CloudCLI gives you two levels of configuration:

• Quick Settings for fast toggles
• Full Settings panel for deeper control.

Quick Settings

Click the gear icon in the top-right corner, and a side panel slides out.

Appearance:

• Dark Mode toggle
• Language selector (English, Japanese, Spanish, and more)

Tool Display:

• Auto-expand tools — Shows tool calls expanded by default
• Show raw parameters — Displays the raw JSON parameters for each tool call
• Show thinking — Reveals the model’s reasoning process

View Options:

• Auto-scroll to bottom — Keeps the chat scrolled to the latest message

Agent Settings

Open the full Settings panel, and you get a comprehensive configuration dashboard.

The left sidebar shows:

• Agents — Manage Claude, Codex, and Gemini connections
• Appearance — Theme and display preferences
• Git — Git configuration
• API & Tokens — API key management
• Tasks — Task management settings
• Plugins — Install and manage plugins
• Notifications — Push notification preferences
• About — Version info

Under Agents, you see tabs for each providerClaude, Codex, and Gemini. Each has sub-tabs for Account, Permissions, and MCP Servers.

Plugin System

CloudCLI supports plugins that add custom tabs with their own frontend and backend.

You install plugins from Git repos directly in Settings > Plugins. Three official plugins are available:

• Project Stats — Shows file counts, lines of code, file-type breakdown, largest files, and recently modified files
• Web Terminal — Full xterm.js terminal with multi-tab support
• CloudCLI Scheduler — Create scheduled prompts that run through Claude Code, Codex, or Gemini CLI on a cron schedule

The plugin architecture is open, and you can fork the Plugin Starter Template and build your own.

Final Thoughts

CloudCLI is a single GUI for managing multiple AI coding agents. After spending the afternoon testing it, here’s what I liked :

• Multi-agent support is the standout feature — It connects Claude Code, Codex, Cursor CLI, and Gemini in one interface.
• Shell tab — Running Claude Code inside the browser means you can access your sessions from any device.
• Source Control — Not having to switch to a separate Git client while working with an AI agent removes friction
• One-command install — npx @cloudcli-ai/cloudcli and you're running. It picks up your existing Claude Code sessions.
• MCP Servers sync both ways — Whatever you configure in CloudCLI writes back to your ~/.claude config and vice versa.

What Could Improve

• Claude Code OAuth expired during my test — I had to re-authenticate mid-session
• The plugin ecosystem is still early — Only three official plugins are available.

If you’re only using Claude Code and want to try different agents in one interface or want remote access to your sessions, CloudCLI is worth the five minutes it takes to set up.

Claude Code Masterclass Course

Every day, I’m working hard to build the ultimate Claude Code course, which demonstrates how to create workflows that coordinate multiple agents for complex development tasks. It’s due for release soon.

It will take what you have learned from this article to the next level of complete automation.

New features are added to Claude Code daily, and keeping up is tough.

The course explores Agents, Hooks, advanced workflows, and productivity techniques that many developers may not be aware of.

Once you join, you’ll receive all the updates as new features are rolled out.

This course will cover:

• Advanced subagent patterns and workflows
• Production-ready hook configurations
• MCP server integrations for external tools
• Team collaboration strategies
• Enterprise deployment patterns
• Real-world case studies from my consulting work

If you’re interested in getting notified when the Claude Code course launches, click here to join the early access list →

**(** Currently, I have 3000+ already signed-up developers)

I’ll share exclusive previews, early access pricing, and bonus materials with people on the list.

Let’s Connect!

If you are new to my content, my name is Joe Njenga

Join thousands of other software engineers, AI engineers, and solopreneurs who read my content daily on Medium and on YouTube where I review the latest AI engineering tools and trends. If you are more curious about my projects and want to receive detailed guides and tutorials, join thousands of other AI enthusiasts in my weekly AI Software engineer newsletter

If you would like to connect directly, you can reach out here:

[embed]AI Automation Software Engineer (10+ Years Experience) My expertise spans LLMs, vector databases, RAG pipelines, MCP servers, and intelligent automation workflows that solve…njengah.com

Follow me on Medium | YouTube Channel | X | LinkedIn | GitHub