GitHub - danielbowne/claude-context: Claude Context with LanceDB local vector database - AI-powered semantic code search for MCP

Your entire codebase as Claude's context

Claude Context is an MCP plugin that adds semantic code search to Claude Code and other AI coding agents, giving them deep context from your entire codebase.

🧠 Your Entire Codebase as Context: Claude Context uses semantic search to find all relevant code from millions of lines. No multi-round discovery needed. It brings results straight into the Claude's context.

💰 Cost-Effective for Large Codebases: Instead of loading entire directories into Claude for every request, which can be very expensive, Claude Context efficiently stores your codebase locally in a vector database and only uses related code in context to keep your costs manageable.

🏠 Local-First with LanceDB: By default, Claude Context uses LanceDB for local, embedded vector storage - no external services required. Quick setup with your data staying on your machine.

🚀 Demo

Model Context Protocol (MCP) allows you to integrate Claude Context with your favorite AI coding assistants, e.g. Claude Code.

🚀 Quick Start

Setup in 30 seconds - Just one command!

Step 1: Get your OpenAI API Key

Get a free API key from OpenAI (starts with sk-)

Step 2: Add to Claude Code

claude mcp add claude-context -e OPENAI_API_KEY=your-openai-api-key -- npx @dannyboy2042/claude-context-mcp@latest

That's it! 🎉 Claude Context will:

✅ Install automatically with no setup required
✅ Store everything locally using LanceDB (no cloud services needed)
✅ Work with any codebase size
✅ Enable hybrid vector + text search for better results

Step 3: Start using it

Open Claude Code in your project directory
Ask Claude to "index this codebase"
Then ask questions like "find the authentication logic" or "show me the database connection code"

📁 Multiple Projects Support

Claude Context automatically handles multiple projects by creating isolated collections for each codebase:

Each project gets its own collection based on the project's absolute path
Projects are completely isolated from each other
Switch between projects by opening Claude Code in different directories
All data stored locally in ~/.claude-context/lancedb/

System Requirements:

Node.js >= 20.0.0 (Claude Context works locally - no external services required!)

See the Claude Code MCP documentation for more details.

Other MCP Client Configurations

Simple setup for any MCP client - just add your OpenAI API key!

Gemini CLI

Create or edit the ~/.gemini/settings.json file:

{
  "mcpServers": {
    "claude-context": {
      "command": "npx",
      "args": ["@dannyboy2042/claude-context-mcp@latest"],
      "env": {
        "OPENAI_API_KEY": "your-openai-api-key"
      }
    }
  }
}

Qwen Code

Create or edit the ~/.qwen/settings.json file:

{
  "mcpServers": {
    "claude-context": {
      "command": "npx",
      "args": ["@dannyboy2042/claude-context-mcp@latest"],
      "env": {
        "OPENAI_API_KEY": "your-openai-api-key"
      }
    }
  }
}

Cursor

Go to: Settings -> Cursor Settings -> MCP -> Add new global MCP server

Add this to your Cursor ~/.cursor/mcp.json file:

{
  "mcpServers": {
    "claude-context": {
      "command": "npx",
      "args": ["-y", "@dannyboy2042/claude-context-mcp@latest"],
      "env": {
        "OPENAI_API_KEY": "your-openai-api-key"
      }
    }
  }
}

Void

Go to: Settings -> MCP -> Add MCP Server

Add the following configuration to your Void MCP settings:

{
  "mcpServers": {
    "code-context": {
      "command": "npx",
      "args": ["-y", "@dannyboy2042/claude-context-mcp@latest"],
      "env": {
        "OPENAI_API_KEY": "your-openai-api-key",
      }
    }
  }
}

Claude Desktop

Add to your Claude Desktop configuration:

{
  "mcpServers": {
    "claude-context": {
      "command": "npx",
      "args": ["@dannyboy2042/claude-context-mcp@latest"],
      "env": {
        "OPENAI_API_KEY": "your-openai-api-key",
      }
    }
  }
}

Windsurf

Windsurf supports MCP configuration through a JSON file. Add the following configuration to your Windsurf MCP settings:

{
  "mcpServers": {
    "claude-context": {
      "command": "npx",
      "args": ["-y", "@dannyboy2042/claude-context-mcp@latest"],
      "env": {
        "OPENAI_API_KEY": "your-openai-api-key",
      }
    }
  }
}

VS Code

The Claude Context MCP server integrates directly with Claude Code and other MCP-compatible clients:

{
  "mcpServers": {
    "claude-context": {
      "command": "npx",
      "args": ["-y", "@dannyboy2042/claude-context-mcp@latest"],
      "env": {
        "OPENAI_API_KEY": "your-openai-api-key",
      }
    }
  }
}

Cherry Studio

Cherry Studio allows for visual MCP server configuration through its settings interface. While it doesn't directly support manual JSON configuration, you can add a new server via the GUI:

Navigate to Settings → MCP Servers → Add Server.
Fill in the server details:
- Name: claude-context
- Type: STDIO
- Command: npx
- Arguments: ["@dannyboy2042/claude-context-mcp@latest"]
- Environment Variables:
  - OPENAI_API_KEY: your-openai-api-key
  - MILVUS_ADDRESS: milvus-cloud-public-endpoint
  - MILVUS_TOKEN: milvus-cloud-api-key
Save the configuration to activate the server.

Cline

Cline uses a JSON configuration file to manage MCP servers. To integrate the provided MCP server configuration:

Open Cline and click on the MCP Servers icon in the top navigation bar.
Select the Installed tab, then click Advanced MCP Settings.
In the cline_mcp_settings.json file, add the following configuration:

{
  "mcpServers": {
    "claude-context": {
      "command": "npx",
      "args": ["@dannyboy2042/claude-context-mcp@latest"],
      "env": {
        "OPENAI_API_KEY": "your-openai-api-key",
      }
    }
  }
}

Save the file.

Augment

To configure Claude Context MCP in Augment Code, you can use either the graphical interface or manual configuration.

A. Using the Augment Code UI

Click the hamburger menu.
Select Settings.
Navigate to the Tools section.
Click the + Add MCP button.

Enter the following command:

npx @dannyboy2042/claude-context-mcp@latest

Name the MCP: Claude Context.
Click the Add button.

B. Manual Configuration

Press Cmd/Ctrl Shift P or go to the hamburger menu in the Augment panel
Select Edit Settings
Under Advanced, click Edit in settings.json
Add the server configuration to the mcpServers array in the augment.advanced object

"augment.advanced": { 
  "mcpServers": [ 
    { 
      "name": "claude-context", 
      "command": "npx", 
      "args": ["-y", "@dannyboy2042/claude-context-mcp@latest"],
      "env": {
        "OPENAI_API_KEY": "your-openai-api-key",
      }
    }
  ]
}

Roo Code

Roo Code utilizes a JSON configuration file for MCP servers:

Open Roo Code and navigate to Settings → MCP Servers → Edit Global Config.
In the mcp_settings.json file, add the following configuration:

{
  "mcpServers": {
    "claude-context": {
      "command": "npx",
      "args": ["@dannyboy2042/claude-context-mcp@latest"],
      "env": {
        "OPENAI_API_KEY": "your-openai-api-key",
      }
    }
  }
}

Save the file to activate the server.

Other MCP Clients

The server uses stdio transport and follows the standard MCP protocol. It can be integrated with any MCP-compatible client by running:

npx @dannyboy2042/claude-context-mcp@latest

For more detailed MCP environment variable configuration, see our Environment Variables Guide.

📚 Need more help? Check out our complete documentation for detailed guides and troubleshooting tips.

🏗️ Architecture

🔧 Key Features

🔍 Hybrid Search: Advanced hybrid search combining vector similarity + full-text search using RRF (Reciprocal Rank Fusion) for better results
🏠 100% Local: Everything runs locally using LanceDB - your code never leaves your machine
⚡ Fast & Incremental: Smart indexing that only processes changed files
🧩 AST-Powered: Intelligent code chunking using Abstract Syntax Trees for better context
🗄️ No Limits: Handle codebases of any size locally
🛠️ Zero Config: Works out of the box - just add your OpenAI API key

Core Components

Claude Context is a monorepo containing three main packages:

@dannyboy2042/claude-context-core: Core indexing engine with embedding and vector database integration
@dannyboy2042/claude-context-mcp: Model Context Protocol server for AI agent integration

Supported Technologies

Vector Database: LanceDB (local, embedded, zero-config)
Embedding: OpenAI text-embedding-3-small (other providers available: VoyageAI, Ollama, Gemini)
Code Analysis: AST-based intelligent chunking with LangChain fallback
Languages: TypeScript, JavaScript, Python, Java, C++, C#, Go, Rust, PHP, Ruby, Swift, Kotlin, Scala, Markdown
AI Clients: Claude Code, Cursor, Windsurf, Gemini CLI, and all MCP-compatible clients

🔧 Advanced: Optional Cloud Vector Database

For enterprise teams or advanced use cases, you can optionally configure Milvus/Zilliz Cloud by setting environment variables:

claude mcp add claude-context -e OPENAI_API_KEY=your-key -e VECTOR_DB_TYPE=milvus -e MILVUS_TOKEN=your-token -- npx @dannyboy2042/claude-context-mcp@latest

📦 Other Ways to Use Claude Context

The MCP server is the primary way to use Claude Context with AI assistants like Claude Code.

Build Applications with Core Package

The @dannyboy2042/claude-context-core package provides the fundamental functionality for code indexing and semantic search.

import { Context, LanceDBVectorDatabase, OpenAIEmbedding } from '@dannyboy2042/claude-context-core';

// Initialize embedding provider
const embedding = new OpenAIEmbedding({
    apiKey: process.env.OPENAI_API_KEY || 'your-openai-api-key',
    model: 'text-embedding-3-small'
});

// Initialize vector database (LanceDB - local, no setup required)
const vectorDatabase = new LanceDBVectorDatabase({
    uri: './.claude-context/lancedb' // Local storage path
});

// Create context instance
const context = new Context({
    embedding,
    vectorDatabase
});

// Index your codebase with progress tracking
const stats = await context.indexCodebase('./your-project', (progress) => {
    console.log(`${progress.phase} - ${progress.percentage}%`);
});
console.log(`Indexed ${stats.indexedFiles} files, ${stats.totalChunks} chunks`);

// Perform semantic search
const results = await context.semanticSearch('./your-project', 'vector database operations', 5);
results.forEach(result => {
    console.log(`File: ${result.relativePath}:${result.startLine}-${result.endLine}`);
    console.log(`Score: ${(result.score * 100).toFixed(2)}%`);
    console.log(`Content: ${result.content.substring(0, 100)}...`);
});

🛠️ Development

Setup Development Environment

# Clone repository
git clone https://github.com/danielbowne/claude-context.git
cd claude-context

# Install dependencies
pnpm install

# Build all packages
pnpm build

# Start development mode
pnpm dev

Building

# Build all packages
pnpm build

# Build specific package
pnpm build:core
pnpm build:vscode
pnpm build:mcp

Running Examples

# Development with file watching
cd examples/basic-usage
pnpm dev

Supported File Extensions

By default, Claude Context supports:

Programming languages: .ts, .tsx, .js, .jsx, .py, .java, .cpp, .c, .h, .hpp, .cs, .go, .rs, .php, .rb, .swift, .kt, .scala, .m, .mm
Documentation: .md, .markdown, .ipynb

Ignore Patterns

Common directories and files are automatically ignored:

Build outputs: node_modules/**, dist/**, build/**, out/**, target/**, coverage/**, .nyc_output/**
Version control: .git/**, .svn/**, .hg/**
IDE/Editor files: .vscode/**, .idea/**, *.swp, *.swo
Cache directories: .cache/**, __pycache__/**, .pytest_cache/**
Logs and temporary: logs/**, tmp/**, temp/**, *.log
Environment files: .env, .env.*, *.local
Minified/bundled files: *.min.js, *.min.css, *.bundle.js, *.bundle.css, *.chunk.js, *.map

See FAQ Guide for detailed and customized configuration of supported file extensions and ignore patterns.

📖 Examples

Check the /examples directory for complete usage examples:

Basic Usage: Simple indexing and search example

❓ FAQ

Common Questions:

For detailed answers and more troubleshooting tips, see our FAQ Guide.

🤝 Contributing

We welcome contributions! Please see our Contributing Guide for details on how to get started.

Package-specific contributing guides:

🗺️ Roadmap

AST-based code analysis for improved understanding
Support for additional embedding providers
Agent-based interactive search mode
Enhanced code chunking strategies
Search result ranking optimization

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

🔗 Links

🙏 Thanks

Special thanks to Cheney Zhang and the Zilliz team for creating the original Claude Context project. This fork builds upon their excellent foundation to provide a local-first vector database experience with LanceDB.

Name		Name	Last commit message	Last commit date
Latest commit History 7 Commits
.github		.github
.vscode		.vscode
assets		assets
docs		docs
packages		packages
python		python
.env.example		.env.example
.eslintrc.js		.eslintrc.js
.gitignore		.gitignore
CONTRIBUTING.md		CONTRIBUTING.md
LICENSE		LICENSE
README.md		README.md
package.json		package.json
pnpm-lock.yaml		pnpm-lock.yaml
pnpm-workspace.yaml		pnpm-workspace.yaml
tsconfig.json		tsconfig.json
yarn.lock		yarn.lock

Folders and files

Latest commit

History

Repository files navigation

Your entire codebase as Claude's context

🚀 Demo

🚀 Quick Start

Step 1: Get your OpenAI API Key

Step 2: Add to Claude Code

Step 3: Start using it

📁 Multiple Projects Support

Other MCP Client Configurations

A. Using the Augment Code UI

B. Manual Configuration

🏗️ Architecture

🔧 Key Features

Core Components

Supported Technologies

📦 Other Ways to Use Claude Context

Build Applications with Core Package

🛠️ Development

Setup Development Environment

Building

Running Examples

Supported File Extensions

Ignore Patterns

📖 Examples

❓ FAQ

🤝 Contributing

🗺️ Roadmap

📄 License

🔗 Links

🙏 Thanks

About

Resources

License

Contributing

Uh oh!

Stars

Watchers

Forks

Releases

Packages 0

Uh oh!

Uh oh!

Contributors 1

Languages

Packages