Repomix MCP Server
An MCP server that packs repositories into AI-friendly files using the Repomix tool, supporting local and remote repositories with size estimation and compression. It enables users to efficiently feed entire codebases into AI models while managing context window limits.
README
Repomix MCP Server
A Model Context Protocol (MCP) server that provides access to the repomix tool for packing repositories into AI-friendly files.
Security
- Input paths: The server restricts file access to the directory from which it was started. Any attempts to access files outside this directory (like
/etc/) will be denied. - Output files: All output is written to the system's temporary directory and automatically cleaned up after the contents are returned.
- Remote URLs: Remote repository URLs are still allowed for processing.
Installation
npm install
npm run build
Usage
Claude Code
claude mcp add --scope user repomix node /path/to/repomix-mcp/dist/index.js
Claude Desktop
Add this server to your MCP client configuration in your claude_desktop_config.json:
{
"mcpServers": {
"repomix": {
"command": "node",
"args": ["/path/to/repomix-mcp/dist/index.js"]
}
}
}
Available Tools
Both tools accept the same parameters:
Parameters
| Parameter | Type | Required | Description | Examples |
|---|---|---|---|---|
path |
string | No | Directory path to pack | /path/to/repo |
style |
enum | No | Output format style | xml, markdown, plain |
compress |
boolean | No | Compress output to reduce token count | true, false |
include |
string | No | Files to include (glob pattern) | *.md,*.ts,*.js, *.py, src/**/*.go |
ignore |
string | No | Files to exclude (glob pattern) | *test*,*spec*,dist/**,build/** |
remote |
string | No | Remote repository URL to process | https://github.com/user/repo |
repomix-estimate
Estimate the size of repomix output without retrieving the content. Use this first to check if the output will fit in your context window.
Returns:
- File size in KB/MB
- Estimated token count (~4 characters per token)
- Whether compression is enabled
repomix-estimate output
Repomix output size estimate:
- Size: 5.27 KB (0.01 MB)
- Estimated tokens: ~1,349
- Compression: disabled
Use the repomix tool with these same parameters to retrieve the actual content.
repomix
Pack a repository into a single, AI-friendly file. Returns the contents of the generated file.
Best Practice: Always use repomix-estimate first to check the output size, then use repomix with appropriate parameters (especially compress=true for large repos).
Example usage in Claude:
- First check size:
use repomix-estimate tool - If size is reasonable:
use repomix tool - If too large, try with compression:
use repomix-estimate tool with compress=true - Then retrieve:
use repomix tool with compress=true
Workflow: Always estimate first, then retrieve only if the size fits your needs.
repomix output (first 15 lines)
This file is a merged representation of a subset of the codebase, containing specifically included files, combined into a single document by Repomix.
<file_summary>
This section contains a summary of this file.
<purpose>
This file contains a packed representation of the entire repository's contents.
It is designed to be easily consumable by AI systems for analysis, code review,
or other automated processes.
</purpose>
<file_format>
The content is organized as follows:
1. This summary section
2. Repository information
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。