Kimi Code
MCP server for Claude Code × Kimi K2.5 (256K context) — delegate bulk codebase analysis to Kimi, save 90% on token costs. Session caching, parallel agents, TypeScript.
What is Kimi Code?
Kimi Code is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to mcp server for claude code × kimi k2.5 (256k context) — delegate bulk codebase analysis to kimi, save 90% on token costs. session caching, parallel agents, typescript.
MCP server for Claude Code × Kimi K2.5 (256K context) — delegate bulk codebase analysis to Kimi, save 90% on token costs. Session caching, parallel agents, TypeScript.
This server falls under the Coding Agents category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- MCP server for Claude Code × Kimi K2.5 (256K context) — dele
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx kimi-codeConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use Kimi Code
Kimi Code MCP Server bridges Claude Code and Moonshot AI's Kimi K2.5 model (256K context window), letting Claude delegate bulk codebase analysis tasks to Kimi rather than consuming expensive Claude tokens for large file reads. It exposes tools for deep architectural analysis, quick programming Q&A, session resumption, and cache management — enabling a workflow where Kimi scans thousands of lines of code in a single pass and Claude acts on the findings. Teams doing large-scale security audits, dependency mapping, or refactoring projects can cut token costs by up to 90% by routing read-heavy tasks through this server.
Prerequisites
- Node.js 18+ and npm installed on your machine
- Kimi CLI installed and authenticated (`curl -L code.kimi.com/install.sh | bash && kimi login`)
- A Kimi Code API key set as the KIMICODE_API_KEY environment variable
- Claude Code CLI or another MCP-compatible client installed
- npm package `kimi-mcp-server` (installed globally or via npx)
Install and authenticate the Kimi CLI
The MCP server shells out to the Kimi CLI to perform analysis. Install it and log in before configuring the MCP server.
curl -L code.kimi.com/install.sh | bash
kimi loginSet your Kimi Code API key
Export the KIMICODE_API_KEY environment variable, or add it to your shell profile so it persists across sessions.
export KIMICODE_API_KEY="your-kimi-api-key-here"Install the MCP server package
Install kimi-mcp-server globally from npm so it can be invoked by your MCP client.
npm install -g kimi-mcp-serverAdd the server to your MCP configuration
Add the kimi-code server to your project's .mcp.json or to the global ~/.claude/mcp.json file.
{
"mcpServers": {
"kimi-code": {
"command": "npx",
"args": ["-y", "kimi-mcp-server"],
"env": {
"KIMICODE_API_KEY": "your-kimi-api-key-here"
}
}
}
}Verify the server status
In Claude Code, use the kimi_status tool to confirm the CLI version, authentication, and connectivity are all healthy before running analysis.
Run your first codebase analysis
Ask Claude to delegate a large codebase analysis to Kimi. Claude will invoke kimi_analyze with a 10-minute timeout and use the results to answer your question.
Kimi Code Examples
Client configuration
MCP configuration block for kimi-code using npx, with the API key passed as an environment variable.
{
"mcpServers": {
"kimi-code": {
"command": "npx",
"args": ["-y", "kimi-mcp-server"],
"env": {
"KIMICODE_API_KEY": "your-kimi-api-key-here"
}
}
}
}Prompts to try
Example prompts that route large analysis tasks to Kimi K2.5 while Claude acts on the results.
- "Use Kimi to analyse the entire src/ directory and give me an architecture overview."
- "Delegate a security audit of the auth module to Kimi, then summarise the top 3 vulnerabilities."
- "Ask Kimi to map all dependencies of the payments service, then plan a refactoring strategy."
- "Resume the previous Kimi session and continue the codebase analysis from where we left off."
- "Use kimi_query to quickly answer: what design pattern does this codebase use for dependency injection?"Troubleshooting Kimi Code
kimi_status reports CLI not found or not authenticated
Re-run the Kimi CLI install script (curl -L code.kimi.com/install.sh | bash) and then run 'kimi login'. Ensure the kimi binary is on your PATH by verifying 'which kimi' returns a valid path.
KIMICODE_API_KEY not found error at runtime
Pass the API key directly in the env block of your MCP config JSON rather than relying on shell environment inheritance, as some MCP clients do not forward all shell variables to spawned processes.
kimi_analyze times out on very large repositories
The tool has a 10-minute timeout. For extremely large codebases, scope the analysis to a specific subdirectory rather than the entire repo, or use kimi_resume to continue a previous session in incremental chunks.
Frequently Asked Questions about Kimi Code
What is Kimi Code?
Kimi Code is a Model Context Protocol (MCP) server that mcp server for claude code × kimi k2.5 (256k context) — delegate bulk codebase analysis to kimi, save 90% on token costs. session caching, parallel agents, typescript. It connects AI assistants to external tools and data sources through a standardized interface.
How do I install Kimi Code?
Follow the installation instructions on the Kimi Code GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with Kimi Code?
Kimi Code works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is Kimi Code free to use?
Yes, Kimi Code is open source and available under the MIT license. You can use it freely in both personal and commercial projects.
Kimi Code Alternatives — Similar Coding Agents Servers
Looking for alternatives to Kimi Code? Here are other popular coding agents servers you can use with Claude, Cursor, and VS Code.
Dify
★ 142.2kProduction-ready platform for agentic workflow development.
Ruflo
★ 54.0k🌊 The leading agent orchestration platform for Claude. Deploy intelligent multi-agent swarms, coordinate autonomous workflows, and build conversational AI systems. Features enterprise-grade architecture, self-learning swarm intelligence, RAG integrat
Goose
★ 45.7kan open source, extensible AI agent that goes beyond code suggestions - install, execute, edit, and test with any LLM
Antigravity Awesome Skills
★ 38.3kInstallable GitHub library of 1,400+ agentic skills for Claude Code, Cursor, Codex CLI, Gemini CLI, Antigravity, and more. Includes installer CLI, bundles, workflows, and official/community skill collections.
AgentScope
★ 25.5kBuild and run agents you can see, understand and trust.
Serena
★ 24.5kA coding agent toolkit that provides IDE-like semantic code retrieval and editing tools, enabling LLMs to efficiently navigate and modify codebases using symbol-level operations instead of basic file reading and string replacements.
Browse More Coding Agents MCP Servers
Explore all coding agents servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up Kimi Code in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use Kimi Code?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.