MTContext
Cloud-hosted MCP server that connects AI clients to MetaTrader 4/5 for live prices, account state, indicators, and trades with dry-run safety via natural language.
README
MTContext MCP Quickstart
Connect Claude, Cursor, or any MCP-compatible client to a MetaTrader 4 or 5 terminal — no Python, no local server to run.
MTContext is a cloud-hosted, remote MCP server. There's nothing to clone,
build, or pip install from this repo — this repo exists to document the
connection and point at the real thing. (This is intentional: the server
implementation itself is closed-source, cloud infrastructure, not something
you run yourself.)
How it works
- Sign up (free tier: read-only market data, no card required).
- Drag one Expert Advisor file (
.ex5for MT5,.ex4for MT4 — MT4 support coming soon) onto a chart in your MetaTrader terminal. - Copy your ready-made MCP config snippet from the dashboard and paste it into your client's config.
- Ask your AI client, in plain English, for live prices, account state,
indicators, or — when you explicitly allow it — trades. Every
trade-affecting call defaults to
dry_run: trueuntil you confirm.
Connection
MTContext exposes a single remote Streamable HTTP MCP endpoint per account:
{
"mcpServers": {
"mtcontext": {
"type": "http",
"url": "https://mcp.mtcontext.com/mcp/v1",
"headers": {
"Authorization": "Bearer YOUR_LICENSE_KEY"
}
}
}
}
Get your real YOUR_LICENSE_KEY from your dashboard
after connecting a terminal — see examples/claude_desktop_config.example.json.
What you can ask it
- "What's the current EURUSD bid and ask?"
- "What's my account balance and open positions?"
- "Run RSI and MACD on GBPUSD, H1."
- "Size a position on XAUUSD risking 1% of my equity with a 30-pip stop."
- "Place a dry-run buy on EURUSD, 0.1 lots, and show me what would happen."
Full tool reference: docs.mtcontext.com. Setup guides and troubleshooting: help.mtcontext.com.
Why not a self-hosted MCP server?
Several open-source MetaTrader MCP servers exist and require installing Python and running a local process yourself, usually MT5-only, with no safety layer beyond what you build. MTContext trades that developer-only setup for a one-file install, MT4 and MT5, and a dry-run-by-default safety model — aimed at traders, not just developers who happen to trade.
Links
- mtcontext.com — sign up
- docs.mtcontext.com — full documentation
- help.mtcontext.com — Help Center
- Model Context Protocol
License
The contents of this repository (documentation and examples) are MIT licensed — see LICENSE. This does not cover the MTContext service itself, which is proprietary.
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.