Moonshot AI has been launched Kimi Code CLI, An open supply coding agent that runs contained in the terminal. This software reads and edits code, executes shell instructions, searches for information, and retrieves internet pages. Then select the next step primarily based on the suggestions you obtain. This undertaking is MIT licensed and on GitHub.
Kim Code CLI is the successor to the outdated kimi-cli. The brand new agent is written in TypeScript and distributed through npm. Moonshot AI’s Kim mannequin is able to use. It will also be configured to make use of different appropriate suppliers.
What’s Kimi Code CLI?
Kim Code CLI is an AI agent for software program growth and terminal operations. You may implement new options, repair bugs, and full refactorings. You too can discover unfamiliar codebases and reply architectural questions. Batch file processing, constructing, and working chained assessments are additionally supported.
The execution mannequin is feedback-driven. Brokers plan steps, modify code, run assessments, and report their actions. By default, read-only operations are carried out mechanically. For file edits or shell instructions, the agent first asks for affirmation. This approval stream places dangerous actions beneath developer management.
The CLI itself is free and requires the MIT license. Entry to the mannequin requires a Kimi Code OAuth or Moonshot AI Open Platform API key.
Foremost options
Moonshot lists a number of options aimed toward lengthy, targeted agent periods.
- Single binary distribution. Installs with one command and requires no Node.js setup.
- Fast startup. Based on Moonshot, TUI might be prepared inside milliseconds.
- Devoted TUI. The interface is tailor-made for prolonged agent periods.
- Video enter. Drop a display screen recording or demo clip into the chat.
- AI-native MCP configuration. Add and authenticate the Mannequin Context Protocol server.
/mcp-config. - Subagents for parallel work. Constructed-in dispatch
coder,discoverandplanSubagents in remoted contexts. - life cycle hook. You may run native instructions to manage software calls, audit choices, and set off notifications.
Set up and first run
Two set up paths exist. Official scripts don’t require Node.js to be put in beforehand.
On macOS or Linux, run the set up script.
curl -fsSL https://code.kimi.com/kimi-code/set up.sh | bash
On Home windows, use PowerShell.
irm https://code.kimi.com/kimi-code/set up.ps1 | iexInternational npm set up requires Node.js 24.15.0 or later.
npm set up -g @moonshot-ai/kimi-codeConfirm the binaries and open the undertaking to start out the interactive UI.
kimi --version
cd your-project
kimiAt first startup, sort: /login Contained in the UI. You may select Kim Code OAuth or Moonshot AI Open Platform API key. To execute a single instruction with out utilizing the UI, use: kimi -p "your process". To renew the earlier session, add the next -C.
Utilization instance
- Perceive the undertaking: Request an structure overview and module dependency diagram.
- Implementation of performance: Clarify signatures, choices, and acceptance standards upfront.
- Bug fixes: Enter the signs, steps to breed, and anticipated conduct.
- Writing and refactoring assessments: Extract repeating patterns and run assessments to substantiate conduct.
- One-time automation: Analyze the logs and print the variety of calls with p50 and p99 latency.
- scheduled process: Ask brokers to arrange reminders or recurring checks through cron.
Plan mode is out there from Shift-Tab or kimi --plan. Output the analysis plan earlier than touching the file. For secure batch work, --yolo or /yolo Skip the approval immediate. of /fork The command creates an experimental department that may be deserted. of /compact The command compresses the context and frees the token. For big-scale investigations, the primary agent can dispatch subagents in parallel.
Kim Code CLI comparability
Kim Code CLI joins a number of established terminal coding brokers. The desk beneath compares three of them. Competitor particulars mirror mid-2026 and should change quickly.
| attribute | Kimi Code CLI | claude code | OpenAI Codex CLI | Gemini CLI |
|---|---|---|---|---|
| developer | Moonshot AI | human | OpenAI | |
| backing mannequin | Kimi mannequin | claude mannequin | GPT-5.3-codex | gemini 2.5 professional |
| Language/Runtime | TypeScript | Node.js | rust | TypeScript |
| set up | Script or npm (Node.js ≥ 24.15.0) | Native installer or npm | npm/native | npm single binary |
| MCP assist | sure (/mcp-config) |
sure | sure | sure |
| subagent | sure (coder, discover, plan) |
sure | sure | No (sequential) |
| plan mode | sure (Shift-Tab) |
sure | sure | sure |
| IDE integration | ACP (Zed, JetBrains) | VS Code, JetBrains | VS Code, IDE | VS Code (Code Help) |
| license | Massachusetts Institute of Know-how | personal | open supply | Apache 2.0 |
All 4 brokers assist the mannequin context protocol. They differ in backing fashions, languages, licensing, and orchestration. Kimi Code CLI and Codex CLI each ship with native subagents. Gemini CLI executes duties sequentially with out subagent assist.
Essential factors
- Kim Code CLI is Moonshot AI’s MIT-licensed terminal coding agent.
- It is written in TypeScript and put in through scripts or npm.
- Constructed-in
coder,discoverandplanSubagents run in remoted contexts. - MCP servers are configured conversationally.
/mcp-configIt is not uncooked JSON. - Inherit kimi-cli and migrate current configurations and periods.
Visible clarification of Marktechpost
Kimi Code CLI Information
01/09
Please test GitHub repository here. Additionally, be at liberty to observe us Twitter Remember to affix us 150,000+ ML subreddits and subscribe our newsletter. grasp on! Are you on telegram? You can now also participate by telegram.
Must accomplice with us to advertise your GitHub repository, Hug Face Web page, product launch, webinar, and so forth.? connect with us
Michal Sutter is a knowledge science skilled with a grasp’s diploma in knowledge science from the College of Padova. With a robust basis in statistical evaluation, machine studying, and knowledge engineering, Michal excels at reworking advanced datasets into actionable insights.
