mt5-mcp
Enables AI assistants and traders to connect to a local MetaTrader 5 terminal, fetch historical market data, run strategy backtests, monitor account status, and manage demo trades via natural language, with an interactive web GUI.
README
MetaTrader 5 (MT5) MCP Server & Web GUI Workstation
A high-performance Model Context Protocol (MCP) Server and interactive Web GUI Workstation for MetaTrader 5 on Windows. This project connects AI assistants (Antigravity, Claude Desktop, Cursor, etc.) and quantitative traders directly to their local MT5 installation to:
- Pull historical market data (OHLCV bars across timeframes M1–MN1 and millisecond-precision tick streams) with high-speed Parquet and CSV exports.
- Execute strategy backtests via built-in Python quantitative engines and native MT5 Strategy Tester automation (
terminal64.exe /config:tester.ini). - Inspect demo account status, equity curves, active symbols, market watch quotes, and open positions.
- Interactive Financial Web GUI: Built with TradingView Lightweight Charts, real-time telemetry, Strategy Studio with equity curve visualizer, and account login manager.
- Strict Trading Safety Guard: Ensures demo and live orders can only execute with explicit confirmation.
Architecture Overview
┌────────────────────────────────────────────────────────┐
│ AI Client │
│ (Antigravity / Claude Desktop / Cursor) │
└───────────────────────────┬────────────────────────────┘
│ MCP Protocol (stdio)
┌───────────────────────────▼────────────────────────────┐
│ MT5 MCP Server (Python) │
│ - FastMCP Server (`src/mt5_mcp/server.py`) │
│ - Data Caching & Parquet/CSV Export Engine │
│ - Quantitative Strategy Backtesting Engine │
└─────────────┬────────────────────────────┬─────────────┘
│ │
┌─────────────▼──────────────┐ ┌──────────▼─────────────┐
│ MetaTrader5 Python API │ │ MT5 Native Tester Auto │
│ - copy_rates_range() │ │ - tester.ini builder │
│ - copy_ticks_range() │ │ - terminal64.exe CLI │
│ - account_info() / login()│ │ - Report XML/HTML │
└─────────────┬──────────────┘ └──────────┬─────────────┘
│ │
┌─────────────▼────────────────────────────▼─────────────┐
│ MetaTrader 5 Windows Terminal │
│ (Demo Account Connected) │
└────────────────────────────────────────────────────────┘
Installation & Prerequisites
1. Requirements
- Windows 10 / 11 (64-bit)
- MetaTrader 5 desktop terminal installed
- Python 3.10, 3.11, or 3.12 (64-bit)
uv(recommended) or standardpip
2. Enable Algorithmic Trading in MT5
In your MetaTrader 5 terminal:
- Navigate to Tools -> Options -> Expert Advisors.
- Check Allow algorithmic trading.
- Check Allow DLL imports (if required by custom indicators).
3. Setup Virtual Environment
# Clone the repository
git clone https://github.com/remomein05/mt5-mcp.git
cd mt5-mcp
# Create virtual environment and install dependencies
uv venv --python 3.12 .venv
.venv\Scripts\activate
uv pip install -e ".[dev]"
Web GUI Workstation (Interactive Dashboard)
The repository includes a dedicated financial-grade Web GUI Dashboard with TradingView charting, real-time market watch, strategy backtesting studio, and account management:
# Launch via console script
mt5-mcp-gui
# Or launch via Python module
.venv\Scripts\python -m mt5_mcp.gui --port 8000
Then open http://127.0.0.1:8000 in your browser.
GUI Features:
- 📊 TradingView Lightweight Charts: Smooth candlestick and volume charting with interactive timeframe selector (
M1,M5,M15,M30,H1,H4,D1). - 📈 Strategy Backtest Studio: Configure strategy parameters (
SMA Crossover,RSI Reversal,Bollinger Bands), run simulations, and inspect Equity Growth Curves, KPIs (Sharpe ratio, Max Drawdown %, Win Rate %), and closed trades logs. - 📦 Historical Data Exporter: Instant one-click extraction of candlestick data to Parquet and CSV files.
- 🔍 Market Watch: Real-time symbol exploration with wildcard search (e.g.
*USD*,Crypto*) and live floating spread indicators. - ⚙️ MT5 Connect & Account Login Modal: Seamlessly connect to your active MT5 terminal or switch/log in to any demo account with custom Login ID, Password, Broker/Server, and optional executable path (with Enter key submission and Esc key dismissal).
- 📋 Open Positions & Native MT5 Tester Panel: View open trades and automate native MQL5 Expert Advisors headlessly with batch
.iniconfigurations.
MCP Client Configuration
Claude Desktop / Antigravity / Cursor (mcp_config.json)
Add the MT5 MCP server to your MCP client configuration:
{
"mcpServers": {
"mt5": {
"command": "C:\\Users\\<USER>\\source\\repos\\mt5_mcp\\.venv\\Scripts\\python.exe",
"args": [
"-m",
"mt5_mcp.server"
],
"env": {
"PYTHONUNBUFFERED": "1"
}
}
}
}
Available MCP Tools
1. Connection & Account Management
| Tool | Parameters | Description |
|---|---|---|
mt5_connect |
path, login, password, server |
Initializes connection to MT5 terminal or logs in to a specific demo account. |
mt5_account_summary |
None | Returns demo account balance, equity, leverage, free margin, and profit. |
mt5_terminal_info |
None | Returns MT5 build version, connection state, ping, and data directories. |
2. Market Exploration & Historical Data
| Tool | Parameters | Description |
|---|---|---|
mt5_get_symbols |
group (e.g. *USD*) |
Searches available symbols with wildcard filtering. |
mt5_symbol_details |
symbol (e.g. EURUSD) |
Returns contract specifications (spread, point, digits, swap rates). |
mt5_symbol_current_tick |
symbol |
Retrieves real-time bid, ask, and last price. |
mt5_fetch_historical_bars |
symbol, timeframe, count, date_from, date_to, export_format |
Extracts historical OHLCV data across timeframes (M1 to MN1) with optional Parquet/CSV export. |
mt5_fetch_ticks |
symbol, count, date_from, date_to, export_format |
Extracts millisecond-precision tick streams with optional Parquet/CSV export. |
3. Strategy Testing & Simulation
| Tool | Parameters | Description |
|---|---|---|
mt5_list_backtest_strategies |
None | Lists built-in backtest models (sma_crossover, rsi_reversal, bollinger_bands). |
mt5_run_strategy_backtest |
symbol, timeframe, strategy_name, bars_count, initial_cash, lot_size, params |
Executes backtest on historical MT5 data and returns Sharpe ratio, Max Drawdown %, Win Rate %, Profit Factor, and trade log. |
mt5_run_native_tester |
expert_name, symbol, period, deposit, currency, leverage, model, from_date, to_date, visual |
Automates native MT5 Strategy Tester using generated .ini configs and headless terminal64.exe execution. |
4. Positions & Trading (Guarded)
| Tool | Parameters | Description |
|---|---|---|
mt5_get_open_positions |
symbol (optional) |
Lists open positions on the demo account. |
mt5_get_order_history |
days, symbol (optional) |
Lists closed deals and historical order executions. |
mt5_place_demo_order |
symbol, order_type, volume, sl, tp, explicit_user_approval |
Places demo orders with strict safety guard requiring explicit_user_approval=True. |
Running Tests
Run the comprehensive pytest suite verifying connectivity, historical extraction, quantitative backtesters, and GUI REST APIs:
.venv\Scripts\pytest -v
License
MIT License.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.