Deep Graph

v1.1.2Search & Data Extractionstable

Enables seamless interaction with public or private code graphs from DeepGraph and CodeGPT to perform semantic searches and dependency analysis. It allows developers to explore architectural relationships, retrieve source code, and analyze repository

mcpmodel-context-protocolcode-graphcodegpt
Share:
394
Stars
0
Downloads
0
Weekly
0/5

What is Deep Graph?

Deep Graph is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to enables seamless interaction with public or private code graphs from deepgraph and codegpt to perform semantic searches and dependency analysis. it allows developers to explore architectural relations...

Enables seamless interaction with public or private code graphs from DeepGraph and CodeGPT to perform semantic searches and dependency analysis. It allows developers to explore architectural relationships, retrieve source code, and analyze repository

This server falls under the Search & Data Extraction category on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • Enables seamless interaction with public or private code gra

Use Cases

Perform semantic searches on public and private code graphs.
Analyze repository architecture and dependencies with DeepGraph.
JudiniLabs

Maintainer

LicenseMIT License
Languagejavascript
Versionv1.1.2
UpdatedMay 18, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

NPM

npx -y mcp-code-graph

Manual Installation

npx -y mcp-code-graph

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use Deep Graph

The Deep Graph MCP server (mcp-code-graph) connects AI coding assistants to DeepGraph and CodeGPT code graph indexes, enabling semantic search over repository architecture, direct source code retrieval, dependency analysis, and documentation search through natural language. It exposes seven specialized tools that understand code relationships rather than just text patterns, making it possible for an AI agent to navigate large codebases, understand how components connect, and reason about the impact of proposed changes. It supports both public GitHub repositories and private graphs authenticated via CodeGPT API keys.

Prerequisites

  • Node.js 18+ with npx available
  • An MCP-compatible client such as Claude Code or Claude Desktop
  • For public graphs: a GitHub repository name in the format owner/repository-name
  • For private graphs: a CodeGPT API key (CODEGPT_API_KEY), organization ID, and graph ID from codegpt.co
1

Add the server for a public GitHub repository

For publicly indexed repositories, pass the owner/repository path as the argument to mcp-code-graph. No API key is needed.

claude mcp add "Deep Graph MCP" npx -- -y mcp-code-graph@latest owner/repository-name
2

Add the server for a private CodeGPT graph

For private code graphs, pass your CodeGPT API key, organization ID, and graph ID as positional arguments.

claude mcp add "Deep Graph MCP" npx -- -y mcp-code-graph@latest YOUR_CODEGPT_API_KEY YOUR_ORG_ID YOUR_GRAPH_ID
3

Share the server with a project team

Add the -s project flag to store the MCP configuration in the project's .mcp.json file so all team members share the same code graph connection.

claude mcp add -s project "Deep Graph MCP" npx -- -y mcp-code-graph@latest owner/repository-name
4

Verify the server is registered

List registered MCP servers and inspect the Deep Graph entry to confirm it is active and pointing at the correct repository.

claude mcp list
claude mcp get "Deep Graph MCP"
5

Add to Claude Desktop config (alternative)

If using Claude Desktop rather than Claude Code CLI, add the server block manually to claude_desktop_config.json.

{
  "mcpServers": {
    "deep-graph-mcp": {
      "command": "npx",
      "args": ["-y", "mcp-code-graph@latest", "owner/repository-name"]
    }
  }
}

Deep Graph Examples

Client configuration (public repository)

Connect Claude Desktop to the mcp-code-graph server for a public GitHub repository. Replace owner/repo with the actual repository path.

{
  "mcpServers": {
    "deep-graph-mcp": {
      "command": "npx",
      "args": ["-y", "mcp-code-graph@latest", "vercel/next.js"]
    }
  }
}

Prompts to try

Use these prompts to explore code architecture and relationships once the server is connected.

- "List all available code graphs for this repository"
- "Semantically search for functions related to user authentication in this codebase"
- "Show me the direct dependencies of the DatabaseService class"
- "What would break if I changed the getUserData function in DatabaseService?"
- "Retrieve the source code for the payment processing module"
- "Give me the folder structure of the src directory"

Troubleshooting Deep Graph

nodes-semantic-search returns no results for a public repository

Not all repositories have been indexed by DeepGraph. Check if the repository appears on deepgraph.com or codegpt.co. If it is not indexed, you will need to create a private graph using the CodeGPT platform.

Authentication fails for private graphs with CODEGPT_API_KEY

Ensure the API key, org ID, and graph ID are passed in the correct positional order as arguments to mcp-code-graph. Verify your API key is active in the CodeGPT dashboard at codegpt.co.

npx -y mcp-code-graph@latest is very slow to start

npx downloads the package on first run. Install it globally with 'npm install -g mcp-code-graph' for faster startup, then replace 'npx -y mcp-code-graph@latest' with 'mcp-code-graph' in your config.

Frequently Asked Questions about Deep Graph

What is Deep Graph?

Deep Graph is a Model Context Protocol (MCP) server that enables seamless interaction with public or private code graphs from deepgraph and codegpt to perform semantic searches and dependency analysis. it allows developers to explore architectural relationships, retrieve source code, and analyze repository It connects AI assistants to external tools and data sources through a standardized interface.

How do I install Deep Graph?

Install via npm with the command: npx -y mcp-code-graph. Then add the server configuration to your AI client's JSON config file (e.g., claude_desktop_config.json or .cursor/mcp.json).

Which AI clients work with Deep Graph?

Deep Graph works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is Deep Graph free to use?

Yes, Deep Graph is open source and available under the MIT License license. You can use it freely in both personal and commercial projects.

Browse More Search & Data Extraction MCP Servers

Explore all search & data extraction servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Quick Config Preview

{ "mcpServers": { "deep-graph-mcp-server": { "command": "npx", "args": ["-y", "mcp-code-graph"] } } }

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide →

Ready to use Deep Graph?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides