Skip to main content
The CC GUI desktop app (app name ccgui, also CCGUI / ccgui) is an open-source AI coding desktop application. It puts several command-line AI coding runtimes into one graphical interface, so you can write code, fix bugs, and commit to Git in a chat-style workflow. The current app identifier is com.zhukunpenglinyutong.ccgui. It is built with Tauri 2 + React + Rust and supports macOS / Windows / Linux.
CC GUI is an independent open-source project. It is not affiliated with, endorsed by, or sponsored by Anthropic, OpenAI, Google, or other third-party tool vendors.
Desktop app overview

​
What it can do

  • Switch among Claude Code, Codex, OpenCode, Grok, Kimi, Pi, DeepSeek Harness, and other runtimes in one app.
  • Use @ to cite files, @@ for project memory, @# for notes, / for slash commands, $ for skills, and ! for prompts. Agents are usually chosen from the input toolbar.
  • Built-in file tree, terminal, Git panel, global search, and Plan panel.
  • Manage configured MCP servers, Skills, Agents, the prompt library, and permission prompts.
  • Desktop-only features such as the project knowledge map, project memory, and context ledger.
  • In-app auto-update.

​
Supported runtimes

The MCP / Plugin marketplace entry is still Coming Soon. The README note that Gemini is enabled by default is outdated. Follow the desktop Settings page.

​
How it relates to other clients

CC GUI also has a VS Code extension, a JetBrains plugin, and a mobile client. Chat and provider concepts are similar. The file tree, built-in terminal, project knowledge map, project memory, and multi-workspace management belong to the desktop app only. On configuration: the provider structure in ~/.ccgui/config.json can be shared with the JetBrains client. Sessions and UI state are not guaranteed to stay in sync across clients. This series covers the desktop app only. Sidebar groups: Get started, Guides, Providers / CLI, Advanced, Community. Each engine has its own page. Detection and install share one page.

​
Start here

Download and install

Install the macOS / Windows / Linux package from GitHub Releases.

Prepare your environment

Set up runtimes, API keys, and each engine CLI.

Providers / CLI

Separate pages for Claude, Codex, Kimi, Grok, OpenCode, Pi, and DSH.

Send the first message

Add a project, choose a provider, and finish the first chat.

Learn the interface

Home, chat, file tree, Git, terminal, and Settings.