RepoMind-MCP

RepoMind-MCP

MCP server for repository mapping, dependency analysis, and architecture diagram generation for JavaScript, TypeScript, and Python projects.

Category
访问服务器

README

RepoMind-MCP (v2.0)

npm version License: MIT

An open-source Model Context Protocol (MCP) server that provides advanced repository mapping, local dependency graphing, and architectural diagram generation capabilities. Designed to run locally over stdio transport, it helps AI clients easily understand the codebase structures of JavaScript, TypeScript, and Python projects.

v2.0 Features & Upgrades

1. File Modification-Based Caching Layer

  • Location: src/cache.ts
  • Features: Caches parsed module dependencies in memory. Validates the cache using fast file modification timestamps (mtimeMs via fs.statSync) to prevent redundant file reads. If a file's modification time changes, its cached dependencies are automatically invalidated.

2. Downstream Impact Analysis

  • Location: src/parser.ts
  • Features: Calculates the reverse dependencies (downstream blast radius) of modifying any target source file using breadth-first search (BFS). Useful for planning refactors and discovering cascading compilation or regression risks.

3. MCP Resources API

  • URI: repomind://workspace/graph
  • Features: Exposes the complete serialized workspace dependency graph in JSON format (contains both the direct adjacencyList and reverseAdjacencyList using relative paths).

4. Dynamic MCP Prompts

  • architecture-review: Exposes a structured system architect review template pre-populated with the target directory's real-time ASCII file structure tree.
  • blast-radius-check: Exposes a refactoring checklist prompt template pre-populated with the real-time downstream dependents list and import chains of the target file.

Exposed Tools

repomind_scan_workspace

Recursively walks and maps a workspace directory.

  • Arguments:
    • basePath (string, required): Absolute path of the workspace root to scan.
  • Outputs: JSON summary containing statistics, ignored lists, and a visual ASCII directory tree.

repomind_get_dependencies

Analyzes local imports in source files to construct an in-memory directed dependency graph.

  • Arguments:
    • targetPath (string, required): Absolute path to the file or directory to analyze.

repomind_generate_diagram

Generates a structured, visually organized Mermaid.js architecture diagram.

  • Arguments:
    • scopePaths (array of strings, optional): Specific files or directories to restrict diagram seed files to.
    • workspaceRoot (string, optional): Root directory to resolve imports.
    • diagramType (string, optional): "mermaid" (default).

repomind_impact_analysis (New in v2.0)

Calculates the downstream blast radius (reverse dependents) of modifying a target file.

  • Arguments:
    • targetPath (string, required): Absolute path of the target file.
    • depth (number, optional): Maximum traversal depth (defaults to 3).

Setup & Building

1. Install Dependencies

npm install

2. Build the Project

npm run build

This compiles TypeScript source from src/ to ESM JavaScript files in the dist/ folder, including executable permissions on entry points.

3. Run the Server

To run the server locally over stdio:

npm start

Configuration & Client Integration

Option A: Standard Release (via NPX or Global Install)

For users who install your published package from the public npm registry.

Using NPX (Recommended)

Add the server directly to your Claude Desktop config without installing it beforehand:

{
  "mcpServers": {
    "repomind-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "repomind-mcp"
      ]
    }
  }
}

Global Installation

Install the package globally:

npm install -g repomind-mcp

Then reference it by command name:

{
  "mcpServers": {
    "repomind-mcp": {
      "command": "repomind-mcp",
      "args": []
    }
  }
}

Option B: Local Development Integration

For developing or running the server locally from this repository.

Add this configuration (replace path in args with the absolute path of your compiled dist/index.js file):

{
  "mcpServers": {
    "repomind-mcp": {
      "command": "node",
      "args": ["C:/Users/dipan/Desktop/RepoMind-MCP/dist/index.js"]
    }
  }
}

Testing

A local integration script is provided under scratch/test-server.js which spins up the built server and runs test calls using JSON-RPC. To run the tests:

node scratch/test-server.js

License

MIT

推荐服务器

Baidu Map

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。

官方
精选
JavaScript
Playwright MCP Server

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。

官方
精选
TypeScript
Magic Component Platform (MCP)

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。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

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。

官方
精选
本地
TypeScript
VeyraX

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。

官方
精选
本地
graphlit-mcp-server

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。

官方
精选
TypeScript
Kagi MCP Server

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。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

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。

官方
精选