Advanced MCP Server
An implementation of the Model Context Protocol (MCP) server that provides AI coding agents with a secure, flexible, and extensible environment for executing coding tasks.
Features
Session Management
Resource Management
Sandboxed Tool Execution
Policy Engine
Multi-Agent Collaboration
Extensible Tooling
Rate Limiting
Custom Tools (Code Analysis, Testing, Documentation)
Getting Started
Prerequisites
Node.js 18+
Docker (optional, for enhanced sandboxing)
Docker Compose (optional)
Installation
Clone the repository
Install dependencies:
npm install
Development
To start the development server:
To start with Docker:
Building
To build the TypeScript code:
Testing
To run tests:
To run tests with coverage:
API Documentation
Detailed API documentation is available in the following formats:
Using with Qwen Code
The MCP Server is designed to work seamlessly with Qwen Code. After starting the server:
Start the server:
npm run devConfigure Qwen Code to use the MCP server by creating a
.qwen/settings.json
file:{ "mcpServers": { "local-fullstack-mcp": { "name": "Local Fullstack MCP Server", "transport": "http", "url": "http://localhost:8080", "default": true } } }Initialize a session:
curl -X POST http://localhost:8080/session/init \ -H "Content-Type: application/json" \ -d '{"tools": ["readFile", "writeFile", "runCommand", "listFiles"]}'Use the session ID with Qwen Code commands to perform operations in a secure, sandboxed environment.
See the Qwen CLI Integration Guide for detailed instructions.
Architecture
The MCP server follows a modular architecture with the following components:
MCP Gateway - Accepts connections (gRPC + WebSocket for streaming)
Session Manager - Handles authentication, session lifecycle, and capability negotiation
Resource Manager - File system abstraction with policy enforcement
Execution Manager - Runs commands/tools inside sandboxed runtimes (Docker)
Policy Engine - Enforces access rules and maintains audit logs
Audit Log - Immutable logging system for accountability
Sandbox Runtime - Docker containers with controlled resources
Workspace Storage - Persistent project storage (bind-mounted or virtual FS)
Security
Sandboxed execution using Docker containers
Resource policy enforcement
Audit logging for all actions
Capability negotiation
Rate limiting
Policy-based access control
License
MIT
This server cannot be installed
hybrid server
The server is able to function both locally and remotely, depending on the configuration or use case.
Provides AI coding agents with a secure, sandboxed environment for executing coding tasks including file operations, command execution, and testing. Features session management, policy enforcement, and Docker-based sandboxing for safe code execution and development workflows.