Kimi Code CLI

Documentation · Issues · 中文

⚠️ This is a personal experimental fork of MoonshotAI/kimi-code. Not affiliated with Moonshot AI. Use at your own risk — do not submit PRs from this fork to upstream.

What's Different in This Fork

Compared to upstream, this fork adds:

• 🌐 i18n / Multi-language support. Complete Chinese-English bilingual support across TUI, CLI, and Web UI. All hardcoded English strings replaced with t() calls. Switch locale via the /settings dialog (aliased as /config), locale selector inside.
• 🤖 Team. Multi-agent discussion and collaboration tool — agents can debate, cross-review, and reach consensus before output.
• 🪟 Windows launchers. start-native.bat launches the native CLI; start-desktop.bat builds/launches a locally vendored desktop shell when present.
• 🥖 Bun as the sole packaging engine. Release binaries are single-file builds via bun build --compile, produced by the CI six-platform matrix (linux/darwin/win32 × x64/arm64). The former default Node.js SEA pipeline has been retired: pi-tui helpers load from the packaged-asset cache, URL-fetch SSRF semantics are identical across runtimes (bundled undici), self-update is engine-aware and still recognizes legacy SEA installs, and /status shows a Runtime row.
• Various other fixes and QoL improvements.

For a deeper, contributor-facing breakdown of these additions and how they integrate with the rest of the project, see DEVELOP.md → "Fork-specific additions vs upstream".

What is Kimi Code CLI

Kimi Code CLI is an AI coding agent that runs in your terminal — it can read and edit code, run shell commands, search files, fetch web pages, and choose the next step based on the feedback it receives. It works out of the box with Moonshot AI’s Kimi models and can also be configured to use other compatible providers.

Install

⚠️ This fork does not publish packages or installers. The official script and npm packages below install upstream MoonshotAI/kimi-code, not this fork — they will not give you the i18n or Team features described above. To run this fork, build it from source (see below). The commands are kept here only to document what upstream's install path looks like.

Upstream install path (installs upstream, not this fork)

• macOS or Linux:

curl -fsSL https://code.kimi.com/kimi-code/install.sh | bash

• Windows (PowerShell):

irm https://code.kimi.com/kimi-code/install.ps1 | iex

On Windows, the CLI has built-in shell detection: it prefers PowerShell 7, then Windows PowerShell, then Git Bash (if installed). To pin a specific shell (for example bash.exe, pwsh.exe, or cmd.exe), set KIMI_SHELL_PATH to its absolute path.

Run This Fork From Source

Requirements: Bun >= 1.4 and Git. Node.js is not required — Bun is the package manager, runtime, and packaging engine here. The Nix build path (flake.nix) is the only thing that still needs Node.js >= 24.15.0.

git clone https://github.com/7723qqq/kimi-code.git
cd kimi-code
git switch norust          # norust is the working branch; origin/HEAD points at it
bun install

Then run the CLI:

bun run dev:local           # build a bundle (~2s) and run the CLI from it
bun run dev:local:watch     # same, rebuilding and re-running on every change

On Windows you can also double-click start-native.bat, which launches the CLI.

💡 dev:cli-only runs src/main.ts directly and works on Bun 1.4.4, which resolves the package.json imports wildcard "#/*": "./src/*.ts". Earlier Bun versions did not, so the script aborted with Cannot find module '#/kimi-harness'; use dev:cli (delegates to scripts/dev.mjs) or dev:local (bundles first) if you hit that.

Quick Start

Open a project and start the interactive UI:

cd your-project
bun run dev:local -- --cwd .

On first launch, run /login inside Kimi Code CLI and choose either Kimi Code OAuth or a Moonshot AI Open Platform API key. After login, try your first task:

Take a look at this project and explain its main directories.

Key Features

• Single-binary distribution. Upstream offers a one-command install with no Node.js setup, PATH gymnastics, or global module conflicts. (This fork does not publish binaries — build from source.)
• Blazing-fast startup. The TUI is ready in milliseconds, so starting a session never feels heavy.
• Purpose-built TUI. A carefully tuned interface, optimized end to end for long, focused agent sessions.
• Video input. Drop a screen recording or demo clip into the chat and let the agent watch what is hard to describe in words — turn a reference clip into a LUT, a long video into a short, a screen recording into working code, and more.
• AI-native MCP configuration. Add, edit, and authenticate Model Context Protocol servers conversationally with /mcp-config, without hand-editing JSON.
• Rich plugin ecosystem. Install skills, MCP servers, and data sources from the marketplace or any GitHub repo, with each install's trust level surfaced up front.
• Subagents for focused, parallel work. Dispatch built-in coder, explore, and plan subagents in isolated contexts while keeping the main conversation clean.
• Lifecycle hooks. Run local commands at key points to gate risky tool calls, audit decisions, trigger desktop notifications, or connect to your own automation.
• Editor & IDE integration (ACP). Drive a Kimi Code CLI session straight from Zed, JetBrains, or any Agent Client Protocol client with kimi acp.

Use it in your editor (ACP)

Kimi Code CLI speaks the Agent Client Protocol, so ACP-compatible editors and IDEs (Zed, JetBrains, …) can drive a session over stdio. Log in once, then point your editor at the kimi acp subcommand — no extra login needed.

Running from source? There is no globally installed kimi binary, so point your editor at the bundle this fork produces. Build it once with bun run dev:local (or bun run build), then reference apps/kimi-code/dist/main.mjs in place of the kimi command below.

For Zed, add this to ~/.config/zed/settings.json:

{
  "agent_servers": {
    "Kimi Code CLI": {
      "type": "custom",
      "command": "kimi",
      "args": ["acp"],
      "env": {}
    }
  }
}

Then open a new conversation in Zed's Agent panel. See Using in IDEs for JetBrains setup and troubleshooting, and the kimi acp reference for the full capability matrix.

Docs

• Getting Started
• Interaction and approvals
• Sessions
• Using in IDEs (ACP)
• Configuration
• Command reference

Develop

bun install

bun run dev:local     # run the CLI from a fresh bundle (use this, not dev:cli)
bun run dev:local:watch
bun run build         # build all workspace packages
bun run typecheck     # TypeScript check (builds packages first)
bun run lint          # oxlint --type-aware
bun run test          # run tests (use `bun run test` to run them under Bun)

See DEVELOP.md for the full command reference and CONTRIBUTING.md for the contribution guide.

Community

• Issues
• For security vulnerabilities, see SECURITY.md.

Acknowledgements

Our TUI is built on top of pi-tui. We thank the authors of pi-tui for their valuable work.

License

Released under the MIT License.