yukkit-e2b-mcp-server 0.3.0

E2B MCP Server

E2B MCP Server (Python)

A production-grade Model Context Protocol server for executing code in secure, isolated E2B sandboxes.

✨ Features

• 🚀 Production-Ready: Comprehensive error handling, logging, and resource management
• 🔒 Secure Isolation: Each sandbox runs in a separate, secure E2B environment
• 🎯 Type-Safe: Full type annotations and Pydantic validation
• 📊 Observable: Detailed logging and sandbox statistics
• 🧹 Auto Cleanup: Automatic resource cleanup and graceful shutdown
• 📦 Modular Design: Clean, maintainable code structure
• ⚡ Easy to Use: Simple API with comprehensive documentation

🛠️ Available Tools

The server provides 9 powerful tools for sandbox interaction:

Tool	Description
create_sandbox	Create a new E2B sandbox with configurable timeout
run_code	Execute Python code using Jupyter notebook syntax
run_command	Run shell commands in the sandbox
read_file	Read file contents from the sandbox
write_file	Write content to files in the sandbox
list_files	List files in a sandbox directory
get_sandbox_url	Get URL for accessing sandbox on specific port
get_file_download_url	Get download URL for a file in the sandbox
kill_sandbox	Terminate and cleanup a sandbox

📋 Requirements

• Python 3.10 or higher
• E2B API key (Get one here)

🚀 Quick Start

Installation

# Install with uv (recommended)
uv install

# Or with pip
pip install -e .

Configuration

1. Get your E2B API key from e2b.dev
2. Set up environment variables:

# Copy example config
cp .env.example .env

# Edit .env and add your API key
E2B_API_KEY=your_api_key_here

Running the Server

Standalone

# Run directly
python -m e2b_mcp_server

# Or programmatically
python -c "from e2b_mcp_server import main; import asyncio; asyncio.run(main())"

With Claude Desktop

Add to your Claude Desktop config:

MacOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "e2b-mcp-server": {
      "command": "uvx",
      "args": ["e2b-mcp-server"],
      "env": {
        "E2B_API_KEY": "your_api_key_here"
      }
    }
  }
}

⚙️ Configuration

Environment variables (see .env.example):

Variable	Default	Description
E2B_API_KEY	(required)	Your E2B API key
LOG_LEVEL	INFO	Logging level (DEBUG, INFO, WARNING, ERROR)
MAX_ACTIVE_SANDBOXES	10	Maximum concurrent sandboxes

Advanced Configuration

The server uses these constants (can be modified in constants.py):

• DEFAULT_SANDBOX_TIMEOUT_MS: 300000 (5 minutes)
• MAX_SANDBOX_TIMEOUT_MS: 3600000 (1 hour)

🏗️ Architecture

Module Structure

e2b_mcp_server/
├── constants.py      # Configuration constants
├── exceptions.py     # Custom exception classes
├── schemas.py        # Pydantic validation schemas
├── utils.py          # Utility functions
├── manager.py        # Sandbox lifecycle management
├── handlers.py       # Tool handler implementations
├── server.py         # MCP server setup and routing
└── __init__.py       # Package exports

Design Principles

• Separation of Concerns: Each module has a single, clear responsibility
• Type Safety: Full type hints and Pydantic validation
• Error Handling: Comprehensive exception hierarchy
• Resource Management: Automatic cleanup and lifecycle tracking
• Testability: Modular design enables easy unit testing

Key Components

SandboxManager (manager.py)

Manages E2B sandbox lifecycle:

• Creates and tracks sandboxes
• Enforces resource limits
• Handles cleanup on shutdown

Tool Handlers (handlers.py)

Implements business logic for each tool:

• Validates inputs using schemas
• Interacts with sandboxes
• Returns standardized responses

Schemas (schemas.py)

Pydantic models for validation:

• Type checking
• Field validation
• Auto-generated JSON schemas

🐛 Debugging

Use the MCP Inspector for debugging:

npx @modelcontextprotocol/inspector \
  uv \
  --directory . \
  run \
  e2b-mcp-server

The Inspector provides a web interface for:

• Testing tool calls
• Viewing request/response logs
• Debugging connection issues

Enable Debug Logging

export LOG_LEVEL=DEBUG
python -m e2b_mcp_server

🧪 Development

Setup Development Environment

# Clone the repository
git clone https://github.com/yukkit/e2b-mcp-server
cd mcp-server/packages/python

# Install dependencies
uv install

# Copy environment config
cp .env.example .env
# Edit .env with your E2B_API_KEY

Project Structure

packages/python/
├── e2b_mcp_server/       # Main package
│   ├── __init__.py
│   ├── server.py         # MCP server
│   ├── manager.py        # Sandbox management
│   ├── handlers.py       # Tool handlers
│   ├── schemas.py        # Validation schemas
│   ├── constants.py      # Configuration
│   ├── exceptions.py     # Custom exceptions
│   └── utils.py          # Utilities
├── tests/                # Test suite (coming soon)
├── examples/             # Usage examples
├── .env.example          # Environment template
├── pyproject.toml        # Package configuration
└── README.md             # This file

Code Style

The codebase follows:

• PEP 8 style guidelines
• Type hints for all functions
• Comprehensive docstrings
• Modular, testable design

Running Tests

# Coming soon
pytest tests/

� Building and Packaging

Quick Build

# Method 1: Using Poetry (recommended)
poetry install
poetry build

# Method 2: Using build.sh script
./build.sh

# Generated files in dist/:
# - e2b_mcp_server-0.1.1-py3-none-any.whl
# - e2b_mcp_server-0.1.1.tar.gz

Local Installation Testing

# Install from built wheel
pip install dist/*.whl

# Test the installation
python -m e2b_mcp_server --help

Publishing to PyPI

# Configure PyPI token
poetry config pypi-token.pypi YOUR_TOKEN

# Publish
poetry publish

# Or build and publish together
poetry publish --build

For detailed packaging instructions, see PACKAGING.md.

�📊 Resource Management

The server includes production-grade resource management:

Sandbox Limits

• Maximum concurrent sandboxes: Configurable (default: 10)
• Automatic limit enforcement
• Clear error messages when limits reached

Automatic Cleanup

• Graceful shutdown handling
• All sandboxes cleaned up on exit
• Temporary sandboxes auto-cleaned after use

Monitoring

# Get current statistics
from e2b_mcp_server.manager import sandbox_manager

stats = sandbox_manager.get_stats()
# Returns: {
#   "active_sandboxes": 3,
#   "max_sandboxes": 10,
#   "sandbox_ids": ["sbx_123", "sbx_456", "sbx_789"]
# }

🔒 Security

• Isolated Execution: Each sandbox runs in isolated E2B environment
• Resource Limits: Configurable timeout and sandbox limits
• API Key Protection: Credentials loaded from environment
• Input Validation: All inputs validated with Pydantic

📄 API Reference

Tool Schemas

All tools use Pydantic schemas for validation. See schemas.py for details.

create_sandbox

{
  "timeoutMs": 300000  # Optional, milliseconds
}

run_code

{
  "code": "print('hello')",      # Required
  "sandboxId": "sbx_..."          # Optional
}

run_command

{
  "command": "ls -la",            # Required
  "sandboxId": "sbx_...",         # Required
  "background": false             # Optional
}

See inline documentation in schemas.py for complete API details.

🤝 Contributing

Contributions are welcome! Please:

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Add tests (when available)
5. Submit a pull request

📝 License

Apache 2.0 - See LICENSE file for details.

🔗 Links

• E2B Documentation
• Model Context Protocol
• GitHub Repository
• Issue Tracker

💬 Support

• 📧 Email: [e-mail]
• 💬 Discord: Join our community
• 🐛 Issues: GitHub Issues

🙏 Acknowledgments

Built with:

• E2B - Secure code execution sandboxes
• MCP - Model Context Protocol
• Pydantic - Data validation
• Python - Programming language

---

Made with ❤️ by the E2B team