IBKR MCP Server
An MCP server for Interactive Brokers, enabling account management, trading operations, and market data queries.
README
IBKR MCP Server
An Interactive Brokers (IBKR) MCP server implementation based on FastMCP 2.0 and MCP StreamableHTTP, providing account management, trading operations, and market data query functionality.
Features
- 🔗 Connection Management: Stable connection with IBKR TWS/Gateway
- 📊 Account Information: Query account summary, positions, and balances
- 💹 Trading Operations: Place orders, cancel orders, query order status
- 📈 Market Data: Real-time and historical market data retrieval
- 🛡️ Type Safety: Data validation using Pydantic
- ⚡ Async Architecture: High-performance asynchronous I/O operations
- 📝 Rich Logging: Structured logging
- 🔧 Flexible Configuration: Support for environment variables and configuration files
Installation
Install from Source
git clone https://github.com/yourusername/ibkr-mcp-server.git
cd ibkr-mcp-server
# Create virtual environment
python3 -m venv venv
source venv/bin/activate
# Install the package
pip install -e .
Development Setup
# Install development dependencies
pip install -e ".[dev]"
# Setup pre-commit hooks
pre-commit install
Quick Start
1. Configure Environment
Create a .env file in the project root:
# MCP Server Settings
MCP__HOST=0.0.0.0
MCP__PORT=8080
# IBKR Connection Settings
IBKR__HOST=127.0.0.1
IBKR__PORT=4002
IBKR__CLIENT_ID=1
IBKR__READONLY=false
# Logging Settings
LOGGING__LEVEL=INFO
2. Test Connection
# Test IBKR connection
python -m ibkr_mcp_server.cli test --host 127.0.0.1 --port 4002
3. Start Server
# Start server
python -m ibkr_mcp_server.cli serve
# Or with custom parameters
python -m ibkr_mcp_server.cli serve --host 0.0.0.0 --port 8080
Configuration
Environment Variables
| Variable | Default | Description |
|---|---|---|
MCP__HOST |
0.0.0.0 |
MCP server listen address |
MCP__PORT |
8080 |
MCP server port |
IBKR__HOST |
127.0.0.1 |
IBKR TWS/Gateway address |
IBKR__PORT |
4002 |
IBKR TWS/Gateway port |
IBKR__CLIENT_ID |
1 |
IBKR client ID |
IBKR__READONLY |
false |
Read-only mode |
LOGGING__LEVEL |
INFO |
Logging level |
IBKR Port Configuration
| Platform | Demo Port | Live Port |
|---|---|---|
| TWS | 4002 | 7496 |
| Gateway | 4002 | 4001 |
MCP Tools
The server provides 9 MCP tools:
Account Management
get_account_summary: Get account summary informationget_positions: Get position information
Trading Operations
place_order: Place an ordercancel_order: Cancel an orderget_open_orders: Get open orders
Market Data
get_market_data: Get real-time market dataget_historical_data: Get historical data
Connection Management
connection_status: Check connection statusreconnect: Reconnect to IBKR
Usage Examples
Place Order
{
"tool": "place_order",
"arguments": {
"contract": {
"symbol": "AAPL",
"sec_type": "STK",
"exchange": "SMART",
"currency": "USD"
},
"order": {
"action": "BUY",
"total_quantity": 100,
"order_type": "LMT",
"lmt_price": 150.0
}
}
}
Get Positions
{
"tool": "get_positions",
"arguments": {}
}
Get Historical Data
{
"tool": "get_historical_data",
"arguments": {
"contract": {
"symbol": "AAPL",
"sec_type": "STK",
"exchange": "SMART",
"currency": "USD"
},
"duration": "1 D",
"bar_size": "1 min"
}
}
Architecture
┌─────────────────────┐
│ MCP Client │
│ (Claude Desktop, │
│ Custom Client) │
└─────────┬───────────┘
│ HTTP/WebSocket
┌─────────┴───────────┐
│ FastMCP Server │
│ (MCP Protocol Layer)│
├─────────────────────┤
│ IBKR MCP Server │
│ (Business Logic) │
├─────────────────────┤
│ IBKR Client │
│ (API Wrapper) │
└─────────┬───────────┘
│ TWS API
┌─────────┴───────────┐
│ TWS/Gateway │
│ (IBKR Platform) │
└─────────────────────┘
Development
Project Structure
src/ibkr_mcp_server/
├── __init__.py # Package initialization
├── server.py # MCP server implementation
├── client.py # IBKR client wrapper
├── models.py # Data models
├── config.py # Configuration management
├── exceptions.py # Exception definitions
└── cli.py # Command line interface
Code Standards
- Use
blackfor code formatting - Use
isortfor import sorting - Use
flake8for code linting - Use
mypyfor type checking
Testing
# Run tests
pytest
# Generate coverage report
pytest --cov=src --cov-report=html
Deployment
Docker Deployment
# Build image
docker build -t ibkr-mcp-server .
# Run container
docker run -p 8080:8080 --env-file .env ibkr-mcp-server
Docker Compose
# Start services
docker-compose up -d
# View logs
docker-compose logs -f
Integration with Claude Desktop
Add to your Claude Desktop MCP configuration:
{
"mcpServers": {
"ibkr": {
"command": "python",
"args": ["-m", "ibkr_mcp_server.cli", "serve"],
"env": {
"IBKR__HOST": "127.0.0.1",
"IBKR__PORT": "4002",
"IBKR__CLIENT_ID": "1"
}
}
}
}
Important Notes
- TWS/Gateway: Ensure IBKR TWS or Gateway is running with API connection enabled
- Port Configuration: Make sure TWS/Gateway API port matches your configuration
- Permissions: Ensure your account has appropriate trading permissions
- Risk Management: Please implement proper risk controls in production environments
- Market Data: Some market data may require subscriptions
Troubleshooting
Common Issues
- Connection Failed: Check if TWS/Gateway is running and API is enabled
- Client ID Conflict: Use different client IDs for multiple connections
- Port Issues: Verify the correct port for your TWS/Gateway setup
- Market Data Errors: Ensure you have proper market data subscriptions
Logging
Enable debug logging for troubleshooting:
LOGGING__LEVEL=DEBUG python -m ibkr_mcp_server.cli serve
License
MIT License - see LICENSE file for details
Contributing
Contributions are welcome! Please feel free to submit a Pull Request.
Support
If you encounter any issues, please file an Issue.
中文文档: README_zh_CN.md
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。