OpenSource-Hub
K

kimi-cli

SHA-256
9.6k stars·Developer Tools·SHA-256 checksum verified

Kimi CLI is a terminal-based AI agent that automates coding, shell execution, and editor integration. Its key advantage over similar tools (e.g., GitHub Copilot CLI) is a built-in shell mode and MCP extensibility.

Smart Download

Download Download Version

v1.49.0 · 91.2 MB

A terminal AI agent that writes code, runs shell commands, and integrates with editors via ACP and MCP, with a unique shell mode toggle.

Core Features

  • Intelligent coding assistant: reads/modifies project code, plans and adjusts execution steps autonomously
  • Shell command mode: press Ctrl-X to toggle between AI and shell modes without leaving the terminal
  • VS Code extension: official plugin for inline assistant with code context awareness
  • ACP protocol integration: works with Zed, JetBrains, and any ACP-compatible editor out of the box
  • Zsh plugin: enables AI agent mode in Zsh via Ctrl-X after simple installation

What It Can't Do

  • Built-in shell mode does not support cd-like commands that change the working directory; use full paths or the Zsh plugin instead
  • This project is being phased out in favor of Kimi Code CLI; installing the new version auto-migrates configs, but the old version remains usable
  • MCP servers require network access; test connectivity after adding them
  • For ACP integration, you must complete login via /login before the agent server can be started

Use Cases

  • Software development: write, debug, refactor code; automate git commands and test execution; analyze crash logs
  • DevOps & sysadmin: perform system management tasks in natural language, like file operations, process monitoring, network diagnostics

Detailed Introduction

Kimi CLI is an AI agent that runs entirely in your terminal, designed to assist with software development, code editing, shell execution, web search, and autonomous task planning. Unlike other CLI-based AI tools (e.g., GitHub Copilot CLI), Kimi CLI offers a native shell mode (Ctrl-X to toggle) so you can run commands without switching contexts. It also integrates deeply with Visual Studio Code via an official extension, supports the Agent Client Protocol (ACP) for use with any ACP-compatible editor (like Zed or JetBrains), and includes MCP (Model Context Protocol) support for managing external tools. The project is actively developed by the MoonshotAI team and is evolving into the more advanced Kimi Code CLI, while remaining fully functional for current users.

Tags

AI AgentCLICode AssistantTerminalMCPACPVS Code ExtensionZsh Integration

Getting Started

1

Download installer

Click the button above to download the installer for your system

2

Install the software

Double-click the downloaded installer and follow the prompts

3

Run pip install kimi-cli (requires Python 3.10+)

4

Launch with kimi, then type /login to authenticate

5

Start interacting with natural language, e.g., 'Write a Python function for Fibonacci'

Install Guide
  1. Run pip install kimi-cli (requires Python 3.10+)
  2. Launch with kimi, then type /login to authenticate
  3. Start interacting with natural language, e.g., 'Write a Python function for Fibonacci'
File Integrity

SHA-256 checksum verified

Checksum extracted from GitHub official Release page

SHA256 Checksum

ee0707da7781915a704285ba597f7085ea8dc369e8a7024e440aad258a4a0e04

This checksum is extracted from the GitHub Release page. Verify file integrity after download.

All SHA-256 checksums on this platform are extracted from the project's official GitHub Release page, without any modification. You can independently verify them on the GitHub Releases page.

Open Source Transparency

View GitHub Source
Environment Guide

Uninstall Info

Run pip uninstall kimi-cli. If you installed the Zsh plugin, remove 'kimi-cli' from your plugins list in ~/.zshrc and delete the directory ~/.oh-my-zsh/custom/plugins/kimi-cli.

No Extra Dependencies

Ready to use after download. No additional runtime required.

Project Info
LicenseApache-2.0
Last Updated2026-07-19T05:30:37Z
GitHub RepositoryOfficial Website

Similar Projects