Skip to main content
Glama
Cloudmeru

MetaTrader 5 MCP Server

by Cloudmeru

MetaTrader 5 MCP Server

MetaTrader 5 integration for Model Context Protocol (MCP). Provides read-only access to MT5 market data through Python commands.

⚡ What's New in v0.5.1

  • Standardized error taxonomy – New helpers define 12 error classes, safe JSON parsing, enum conversion, and field validation so every tool returns consistent, timestamped diagnostics.

  • Tool-level guardrailsmt5_query, mt5_analyze, and execute_mt5 now enforce payload size limits, reject dangerous operations, validate MT5 connectivity up front, and catch-all exceptions without crashing the server.

  • Handler resilience – Request objects, enums, and MT5 parameters are validated before execution, with wrapped exceptions that include operation context for easier debugging.

  • Connection + executor hardening – MT5 initialization retries, namespace fallback messaging, and result-format safety nets keep stdio/HTTP transports stable even when inputs misbehave.

  • Expanded regression tests – 30+ new cases cover malformed JSON, invalid enums, missing fields, MT5 connection failures, and rate-limit handling for production readiness.

# Default behavior (stdio only, backward compatible) python -m mt5_mcp # or simply: mt5-mcp # Streamable HTTP with rate limiting and a custom port python -m mt5_mcp --transport http --host 0.0.0.0 --port 7860 --rate-limit 30 # or: mt5-mcp --transport http --host 0.0.0.0 --port 7860 --rate-limit 30 # Dual mode (stdio + HTTP) python -m mt5_mcp --transport both # or: mt5-mcp --transport both

📖 Documentation:

  • USAGE.md - Comprehensive instructions, tool reference, and troubleshooting

  • CHANGELOG.md - Release history and migration notes

Key Capabilities

  • Read-only MT5 bridge – Safe namespace exposes only data-retrieval APIs and blocks all trading calls.

  • Transaction history access – Retrieve and analyze trading history with history_deals_get, history_orders_get, and positions_get.

  • Multiple interaction models – Write Python (execute_mt5), submit structured MT5 queries (mt5_query), or run full analyses with indicators, charts, and forecasts (mt5_analyze).

  • Technical analysis toolkitta, numpy, and matplotlib ship in the namespace for RSI, MACD, Bollinger Bands, multi-panel charts, and more.

  • Interactive charting – Optional Plotly support (px, go) for creating interactive HTML charts from market data and trading history.

  • Forecasting + ML signals – Prophet forecasting and optional XGBoost buy/sell predictions with confidence scoring.

  • LLM-friendly guardrails – Clear tool descriptions, runtime validation, and result-assignment reminders keep assistant output predictable.

Available Tools

execute_mt5

Free-form Python execution inside a curated namespace. Ideal for quick calculations, prototyping, and bespoke formatting.

rates = mt5.copy_rates_from_pos('BTCUSD', mt5.TIMEFRAME_H1, 0, 100) df = pd.DataFrame(rates) df['RSI'] = ta.momentum.rsi(df['close'], window=14) result = df[['time', 'close', 'RSI']].tail(10)

mt5_query

Structured JSON interface that maps directly to MT5 read-only operations with automatic validation, timeframe conversion, and friendly error messages.

{ "operation": "copy_rates_from_pos", "symbol": "BTCUSD", "parameters": {"timeframe": "H1", "count": 100} }

mt5_analyze

Pipeline tool that chains a query → optional indicators → charts and/or Prophet forecasts (with optional ML signals) in one request.

{ "query": { "operation": "copy_rates_from_pos", "symbol": "BTCUSD", "parameters": {"timeframe": "D1", "count": 180} }, "indicators": [ {"function": "ta.trend.sma_indicator", "params": {"window": 50}}, {"function": "ta.momentum.rsi", "params": {"window": 14}} ], "forecast": {"periods": 30, "plot": true, "enable_ml_prediction": true} }

Prerequisites

  • Windows OS (MetaTrader5 library is Windows-only)

  • MetaTrader 5 terminal installed and running

  • Python 3.10+

Installation

  1. Clone this repository:

git clone <repository-url> cd MT5-MCP
  1. Install the package:

pip install -e .

Optional features (HTTP transport and interactive charting):

# Install with HTTP/Gradio support pip install -e .[ui] # Install with Plotly for interactive charts pip install -e .[charting] # Install everything pip install -e .[all] # From PyPI: pip install "mt5-mcp[all]"

Package naming: The package is named mt5-mcp (with hyphen) on PyPI, but the Python module uses mt5_mcp (with underscore). This is standard Python convention.

This will install all required dependencies:

  • mcp - Model Context Protocol SDK

  • MetaTrader5 - Official MT5 Python library

  • pandas - Data manipulation and formatting

  • prophet - Time series forecasting

  • xgboost - Machine learning for trading signals (NEW!)

  • scikit-learn - ML utilities and preprocessing (NEW!)

  • ta - Technical analysis indicators

Configuration

Claude Desktop

Add to your Claude Desktop configuration file:

Windows: %APPDATA%\Claude\claude_desktop_config.json

{ "mcpServers": { "mt5": { "command": "python", "args": ["-m", "mt5_mcp"] } } }

Alternatively, if the mt5-mcp CLI script is in your PATH:

{ "mcpServers": { "mt5": { "command": "mt5-mcp", "args": [] } } }

With Logging (for troubleshooting)

{ "mcpServers": { "mt5": { "command": "python", "args": ["-m", "mt5_mcp", "--log-file", "C:\\path\\to\\mt5_mcp.log"] } } }

Transport Modes (CLI)

Choose how the server exposes MCP transports directly from the command line:

# Default behavior (run only stdio like previous version) python -m mt5_mcp # or: mt5-mcp # Run both transports python -m mt5_mcp --transport both # or: mt5-mcp --transport both # Run only streamable HTTP python -m mt5_mcp --transport http --host 0.0.0.0 --port 7860 # or: mt5-mcp --transport http --host 0.0.0.0 --port 7860

Additional flags:

  • --rate-limit <value> – Requests per IP each minute (set to 0 to disable; keep enabled for public servers).

  • --log-level / --log-file – Tailored diagnostics across transports.

HTTP MCP Clients

  1. Install the optional extras: pip install "mt5-mcp[ui]" (or pip install -e .[ui] while developing).

  2. Launch the HTTP transport: python -m mt5_mcp --transport http --host 0.0.0.0 --port 7860 (or mt5-mcp --transport http).

  3. Point any MCP client to the new endpoint:

{ "mcpServers": { "mt5-http": { "url": "http://localhost:7860/gradio_api/mcp/" } } }

Note: Use python -m mt5_mcp (module name with underscore) or the mt5-mcp CLI command (package name with hyphen) interchangeably.

This endpoint works with MCP Inspector, Claude Desktop (when configured for HTTP), VS Code extensions, or remote deployments (Hugging Face Spaces, Windows VPS, etc.).

Usage Overview

Refer to USAGE.md for a complete walkthrough that covers prerequisites, configuration screens, troubleshooting tips, and in-depth per-tool examples. Below is a quick multi-line example using execute_mt5:

from datetime import datetime, timedelta end_date = datetime.now() start_date = end_date - timedelta(days=30) rates = mt5.copy_rates_range('EURUSD', mt5.TIMEFRAME_D1, start_date, end_date) df = pd.DataFrame(rates) df['time'] = pd.to_datetime(df['time'], unit='s') df['return'] = df['close'].pct_change() result = df[['time', 'close', 'return']].tail(10)

Note: Always assign the final output to result (or another variable noted in USAGE.md) so the MCP response can be formatted correctly.

Architecture & Compliance

  • Built on mcp.server.lowlevel.Server for stdio clients and Gradio v6 for streamable HTTP/SSE, both sharing the same MT5-safe namespace.

  • Safe execution namespace exposes vetted objects (mt5, datetime, pd, ta, numpy, matplotlib) while blocking trading calls and disallowed modules.

  • Runtime validation catches mt5.initialize() / mt5.shutdown() attempts, highlights the correct workflow, and enforces result assignment.

  • Thread-safe MT5 connection management plus IP-scoped rate limiting protect terminals from abusive HTTP workloads.

  • Documentation, tool signatures, and CLI examples match MCP SDK and Gradio MCP guidance for predictable LLM behavior.

Troubleshooting

MT5 Connection Issues

  1. Ensure MT5 terminal is running before starting the MCP server

  2. Enable algo trading in MT5: Tools → Options → Expert Advisors → Check "Allow automated trading"

  3. Check MT5 terminal logs for any errors

Enable Logging

Run the server with logging enabled:

python -m mt5_mcp --log-file mt5_debug.log # or: mt5-mcp --log-file mt5_debug.log

Or configure it in Claude Desktop config (see Configuration section above).

Common Errors

"MT5 connection error: initialize() failed"

  • MT5 terminal is not running

  • MT5 is not installed

  • Algo trading is disabled in MT5

"Symbol not found"

  • Check symbol name spelling (case-sensitive)

  • Symbol may not be available in your MT5 account

  • Use mt5.symbols_get() to see available symbols

"No data returned"

  • Symbol may not have historical data for requested period

  • Check date range validity

  • Some symbols may have limited history

Security

This server provides read-only access to MT5 data. Trading functions are explicitly excluded from the safe namespace:

Blocked Functions

  • order_send() - Place orders

  • order_check() - Check order

  • positions_get() - Get positions (read-only but blocked to prevent confusion)

  • positions_total() - Position count

  • All order/position modification functions

Only market data and information retrieval functions are available.

License

MIT License

Contributing

Contributions are welcome! Please ensure:

  1. All code follows the read-only philosophy

  2. Tests pass (when test suite is added)

  3. Documentation is updated

-
security - not tested
A
license - permissive license
-
quality - not tested

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/Cloudmeru/MetaTrader-5-MCP-Server'

If you have feedback or need assistance with the MCP directory API, please join our Discord server