MCP server connecting Claude Code with Kimi K2.5 (256K context) to delegate bulk codebase analysis and save up to 90% on token costs.
kimi-code-mcp is worth checking the docs before setup with trust notes worth reviewing. Check agent compatibility and use-case fit before adding it to your workflow.
gh repo view howardpen9/kimi-code-mcp --webOpen the official README and confirm the supported install method.
Add the server entry to your MCP client config.
Restart your agent and verify that the server tools appear.
This tool lets Claude Code use Kimi K2.5 to read and analyze large codebases. Instead of paying Claude to read many files, you send the work to Kimi which is much cheaper. It includes session caching and parallel agents for faster analysis.
kimi-code-mcp is an MCP (Model Context Protocol) server that bridges Claude Code with Kimi Code (powered by Kimi K2.5, a 1T MoE model with 256K token context). It allows Claude to orchestrate tasks while Kimi handles heavy reading and analysis of codebases, reducing token costs by up to 90%. The server provides 7 tools for code analysis, including kimi_analyze which can replace 50+ file reads. Features include session caching, parallel agent spawning, and full developer toolchain access (shell, file, web). It requires an active Kimi Code subscription and the Kimi CLI installed. The server is written in TypeScript and uses the official MCP SDK.
Looks usable, but maintenance, license, or security notes deserve a closer look.
Last commit was about 73 days ago.
50 GitHub stars indicate community interest.
1 open issues signal maintenance load.
MIT license detected.
Analyze entire codebases in one pass using Kimi's 256K context window
Delegate bulk file reading and code scanning to Kimi while Claude focuses on reasoning and edits
Perform parallel code analysis tasks with multiple agents
Reduce API costs by offloading token-heavy operations to a cheaper model
Integrate Kimi's code intelligence into Claude Code workflows
Requires an active Kimi Code subscription and API key; costs may apply
Sends code to Kimi's servers for analysis; review data privacy policies
The MCP server runs locally but communicates with external APIs
50
Stars
7
Forks
1
Issues
MIT
License
Local codebase intelligence CLI and MCP server for AI coding agents with change-safety gates and audit evidence.
An offline MCP server that indexes your codebase for semantic search, code search, and git history retrieval.
Official MCP reference servers from Anthropic. Includes servers for filesystem, GitHub, Postgres, Slack, and more.
3 security/trust notes recorded.
Setup difficulty is 3/5.