Codebase MCP Server
A server for secure and efficient codebase analysis.
Codebase MCP Server
Model Context Protocol server for secure and efficient Codebase analysis
Key Features • Supported Languages • Tools • QuickStart • Build • Contributing
🌟 Key Features
- Secure Access: Restricts file operations to predefined root directory.
- Efficient File Management: Provides tools for reading and searching files.
- Detailed Metadata: Retrieves comprehensive file metadata including size, creation time, last modified time, permissions, and type.
- Dependency Analysis: Traverses and analyzes dependency trees within projects.
📋 Supported Languages
- JavaScript/TypeScript
- CSS / CSS Preprocessors
🛠️ Tools
get-project-basics
: Retrieves essential project information including package.json details, directory structure.search-config-files
: Searches for configuration files within the root directory and returns their paths.get-dependency-tree
: Traverses the dependency tree based on the given file path and root directory, and returns the traversal results.list-directory
: Lists the contents of a specified directory, distinguishing between files and directories.read-file-with-metadata
: Reads the content of a specified file and retrieves its metadata.
🚀 QuickStart
Prerequisites
- Node.js v18 or later
- Codebase to communicate with LLM
Installation
First, install the Codebase MCP server with your client. A typical configuration looks like this:
{ "mcpServers": { "Codebase": { "command": "npx", "args": ["codebase-mcp-server@latest", "/path/to/your/codebase"] } } }
Or, you can install the Codebase MCP server with Docker.
After cloning this repository, build the Docker image:
docker build -t mcp/codebase -f .
And then add the following to your MCP servers file:
{ "mcpServers": { "Codebase": { "command": "docker", "args": [ "run", "-i", "--rm", "--mount", "type=bind,src=/path/to/your/codebase/dir,dst=/projects/path/to/your/codebase/dir,ro", "--mount", "type=bind,src=/path/to/some/file.txt,dst=/projects/path/to/some/file.txt", "mcp/codebase", "/projects" ] } } }
Install in VS CodeYou can install the Codebase MCP server using the VS Code CLI:
For VS Code
code --add-mcp '{"name":"Codebase","command":"npx","args":["codebase-mcp-server@latest"]}'
After installation, the Codebase MCP server will be available for use with your GitHub Copilot agent in VS Code.
Install in CursorGo to Cursor Settings -> MCP -> Add new MCP Server. Use following configuration:
{ "mcpServers": { "Codebase": { "command": "npx", "args": ["codebase-mcp-server@latest"] } } }
Install in Windsurf
Follow Windsuff MCP documentation. Use following configuration:
Install in Claude Desktop
Follow the MCP install guide, use following configuration:
🔧 Build
Local Development Build
Install dependencies
npm ci
Build TypeScript
npm run build
Docker Build
Build Docker image
docker build -t mcp/codebase -f .
Or with specific tag
docker build -t mcp/codebase:latest -f .
🤝 Contributing
Contributions are welcome! Please read our Contributing Guide for details on our Code of conduct and the process for submitting pull requests.
Related Servers
Authless Remote MCP Server
An example of a remote MCP server without authentication, deployable on Cloudflare Workers or runnable locally via npm.
BlenderMCP
Connects Blender to AI models via MCP for prompt-assisted 3D modeling, scene creation, and manipulation.
MCP-Portainer Bridge
Manage Docker containers through the Portainer API.
Smithery Reference Servers
A collection of reference implementations for Model Context Protocol (MCP) servers in Typescript and Python, demonstrating MCP features and SDK usage.
Fused MCP
A Python-based MCP server for data scientists to run Python code with a Claude client.
Juniper Junos MCP Server
An MCP server for interacting with Juniper Junos network devices using LLMs.
MCP Prompt Optimizer
Optimize prompts with research-backed strategies for 15-74% performance improvements.
PyMilvus Code Generate Helper
Retrieves relevant code snippets and documents to assist in generating PyMilvus code, requiring a running Milvus instance.
Bazel MCP Server
Exposes the Bazel build system to AI agents, enabling them to build, query, test, and manage dependencies.
Gemini Imagen 3.0
Generate high-quality images using Google's Imagen 3.0 model via the Gemini API.