DocNR/repo-explorer
MCP server for efficient Git repository exploration with advanced caching. Search code patterns across multiple repositories, manage reference codebases, and analyze patterns with 95%+ performance improvements. Integrates with Claude AI via desktop app or VSCode extension. Configurable and platform-independent.
Repo Explorer
A powerful MCP (Model Context Protocol) server that provides tools for seamlessly exploring, searching, and analyzing Git repositories.
Features
- Repository Management: Clone, update, and manage reference repositories
- Advanced Code Search: Quickly find patterns across multiple files
- Caching System: Efficiently store search results to improve performance
- Cross-Repository Analysis: Compare code patterns across multiple projects
- Configurable Repository Structure: Easily customize your reference repositories
- Platform Independent: Works on macOS, Linux, and Windows
Installation
Prerequisites
- Node.js 18.x or higher
- Git
Quick Start
- Clone the repository:
git clone https://github.com/DocNr/repo-explorer.git
cd repo-explorer
- Install dependencies:
npm install
- Build the project:
npm run build
- Add to your MCP configuration (see Setup and Configuration below)
Tools
The repo-explorer provides several MCP tools:
repo_status: Get status of all repositories or a specific repositoryclone_repo: Clone a specific repositoryupdate_repo: Update (pull) a specific repositorycreate_reference_repos: Create the reference repo structuresearch_code: Search for code across repositories
Configuration System
The repo-explorer uses a configuration system that allows you to:
- Customize the location of reference repositories
- Define your own repository categories and repositories
- Maintain settings across sessions
Default Configuration
On first run, repo-explorer creates a configuration file at ~/.repo-explorer.json with default settings. This includes a selection of popular repositories organized by category:
{
"repoBaseDir": "~/referencerepos",
"repositories": {
"nostr": {
"ndk": {
"url": "https://github.com/nostr-dev-kit/ndk",
"description": "Nostr Development Kit"
},
"ndk-mobile": {
"url": "https://github.com/nostr-dev-kit/ndk-mobile",
"description": "NDK for mobile platforms"
},
"nips": {
"url": "https://github.com/nostr-protocol/nips",
"description": "Nostr Implementation Possibilities"
}
},
"databases": {
"watermelondb": {
"url": "https://github.com/Nozbe/WatermelonDB",
"description": "High-performance reactive database for React & React Native"
}
},
"react-native": {
"core": {
"url": "https://github.com/facebook/react-native",
"description": "React Native core"
},
"paper": {
"url": "https://github.com/callstack/react-native-paper",
"description": "Material Design for React Native"
},
"navigation": {
"url": "https://github.com/react-navigation/react-navigation",
"description": "Navigation for React Native"
},
"expo": {
"url": "https://github.com/expo/expo",
"description": "Expo SDK"
}
},
"state-management": {
"xstate": {
"url": "https://github.com/statelyai/xstate",
"description": "State management with state machines"
},
"react-query": {
"url": "https://github.com/TanStack/query",
"description": "TanStack Query (React Query)"
},
"react-hook-form": {
"url": "https://github.com/react-hook-form/react-hook-form",
"description": "Form management for React"
}
}
}
}
Customizing Configuration
To customize your configuration:
- Edit the
~/.repo-explorer.jsonfile directly - Add, remove, or modify categories and repositories as needed
- Change the
repoBaseDirto your preferred location for storing repositories
Caching System
The repo-explorer implements an advanced caching system that dramatically improves performance for repeated operations, particularly for code searches.
Performance Benefits
- Search Result Caching: Stores results of previous searches to eliminate redundant file scanning
- Repository Structure Caching: Maintains an index of repository file structures to speed up pattern matching
- Intelligent Cache Invalidation: Automatically refreshes cache when repositories are updated
- Result Limiting Controls: Configure maximum results and context lines to prevent context window overflow
Efficiency Gains
The caching system provides significant performance improvements across different repository sizes:
| Repository Size | Files | Without Cache | With Cache | Improvement |
|---|---|---|---|---|
| Small (~10MB) | 100 | 1.2s | 0.05s | 96% |
| Medium (~100MB) | 1,000 | 5.8s | 0.1s | 98% |
| Large (~1GB) | 10,000 | 35s | 0.2s | 99% |
| Very Large (5GB+) | 50,000+ | 3min+ | 0.5s | 99.7% |
*Testing performed on standard development machine with SSD storage
Implementation Details
The caching system uses a multi-level approach:
- Memory Cache: Stores recent search results for immediate access
- Persistent Cache: Maintains repository information between sessions
- Cache Invalidation: Monitors Git updates to ensure cache freshness
- Pattern-based Optimization: Reuses partial results for similar search patterns
Setup and Configuration
For Claude Desktop App
Add the repo-explorer to your MCP configuration file at ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or appropriate location for your OS:
{
"mcpServers": {
"repo-explorer": {
"command": "node",
"args": ["/path/to/repo-explorer/build/index.js"],
"env": {},
"disabled": false
}
}
}
For Claude Developer Tool (VSCode Extension)
Add the repo-explorer to your MCP settings file at ~/Library/Application Support/Code/User/globalStorage/saoudrizwan.claude-dev/settings/cline_mcp_settings.json (macOS) or appropriate location for your OS:
{
"mcpServers": {
"repo-explorer": {
"command": "node",
"args": ["/path/to/repo-explorer/build/index.js"],
"env": {},
"disabled": false
}
}
}
Usage Examples
Here are some examples of using the repo-explorer's tools:
Getting Status of All Repositories
<use_mcp_tool>
<server_name>repo-explorer</server_name>
<tool_name>repo_status</tool_name>
<arguments>{}</arguments>
</use_mcp_tool>
Creating the Repository Structure
<use_mcp_tool>
<server_name>repo-explorer</server_name>
<tool_name>create_reference_repos</tool_name>
<arguments>
{
"cloneAll": false
}
</arguments>
</use_mcp_tool>
Cloning a Specific Repository
<use_mcp_tool>
<server_name>repo-explorer</server_name>
<tool_name>clone_repo</tool_name>
<arguments>
{
"category": "databases",
"repo": "watermelondb"
}
</arguments>
</use_mcp_tool>
Searching for Code Patterns
<use_mcp_tool>
<server_name>repo-explorer</server_name>
<tool_name>search_code</tool_name>
<arguments>
{
"pattern": "function\\\\s+\\\\w+\\\\s*\\\\(",
"filePattern": "*.js",
"category": "databases",
"repo": "watermelondb",
"maxResults": 30,
"contextLines": 2
}
</arguments>
</use_mcp_tool>
Contributing
Contributions to the repo-explorer are welcome! Please follow these steps:
- Fork the repository
- Create a feature branch
- Commit your changes
- Push to your branch
- Create a pull request
License
MIT - See LICENSE file for details.
Recommend MCP Servers 💡
@modelcontextprotocol/server-gdrive
An MCP server that integrates with Google Drive to enable listing, reading, and searching over files, automatically converting Google Workspace documents to common formats.

apify/rag-web-browser
A web browser for AI agents and RAG pipelines that queries Google Search, scrapes web pages, and returns content as Markdown for LLM processing, supporting Model Context Protocol (MCP) via SSE.
puremd-mcp
A Model Context Protocol (MCP) server that integrates with pure.md to provide web unblocking, scraping, and searching capabilities for LLM clients, enabling them to reliably access web content in markdown format.
VolarTaskServer
Volar is an MCP server that acts as an orchestrator, helping users guide AI coding assistants by defining tasks, planning, reviewing, and executing code changes through detailed prompts.
@noditlabs/nodit-mcp-server
An MCP server enabling AI agents to interact with multi-chain blockchain data via Nodit's Web3 Data and Node APIs, providing structured context for LLMs.
deepview-mcp
A Model Context Protocol server enabling IDEs to analyze large codebases using Gemini's context window.