Name: Twelve Data
Author: twelvedata

LABSCO SUMMARY

Market data for stocks, forex and crypto on your own key.

A deliberately thin bridge: the assistant asks for a quote, a history or what an instrument is, and the answer comes from the Twelve Data API rather than the model's memory. Coverage and rate limits are whatever your plan already gives you.

What it is

Twelve Data's open-source connector between its market-data API and MCP clients, so an assistant can fetch real-time and historical prices, time series and instrument reference data in a consistent format.

What you get

Quotes, time series and instrument information for stocks, forex pairs and cryptocurrencies
Real-time and historical data from the same Twelve Data API a trading dashboard would call
A backend any MCP-compatible client or desktop app can point at
A local or remote deployment, configured with your own API key

Requirements

A Twelve Data account. The hosted server at https://mcp.twelvedata.com/mcp needs nothing installed: add it as a connector and sign in with your Twelve Data account in the browser, and your API key is linked for you. To run it on your own machine instead, you need Python 3.10+ and your API key from the dashboard's API Keys page: run make install in the checkout, put TWELVE_DATA_API_KEY in a .env file in the project root, and point Claude Desktop at the checkout's .venv Python.

Setup effort

Paste a URL, then authorize — add the endpoint to your client, then approve the OAuth consent screen

See installation options

USE CASES

01

Checking where a price is right now

Ask for the latest quote on a ticker, an FX pair or a coin and get the live figure rather than a remembered one.

02

Pulling a history to reason over

Fetch the time series for the instrument and period you care about, in a shape the model can work through.

03

Looking up what an instrument actually is

Retrieve the reference data for a symbol before building an analysis on top of it.

04

Using one market-data backend across several clients

Point any MCP-compatible desktop app or agent at the same server so they all read the same source.

BEFORE YOU INSTALL

What to know first

Requires

TWELVE_DATA_API_KEY in your environment

Where it runs

Of the 3 documented setups, 1 runs the server on your machine and 2 connect to a hosted endpoint.

INSTALLATION

Connect your client

Server setup

Build from source

TERMINAL

git clone https://github.com/twelvedata/mcp
cd mcp

Log in at twelvedata.com -> your dashboard -> API Keys. Copy your key.

TERMINAL

make install

Create a .env file in the project root with your key:

FILE · .env

TWELVE_DATA_API_KEY=<YOUR_TWELVE_DATA_API_KEY>

Add the mcpServers block to ~/Library/Application Support/Claude/claude_desktop_config.json, then restart Claude Desktop

Hosted endpoint

Claude Desktop, ChatGPT

ENDPOINT URL

https://mcp.twelvedata.com/mcp

Authentication: OAuth — you log in with your Twelve Data account in the browser; no API key to paste

MCP Clients

Claude Desktop — local build

CLAUDE DESKTOP — LOCAL BUILD · JSON

{
  "mcpServers": {
    "twelvedata": {
      "command": "<YOUR_PATH_TO_MCP>/.venv/bin/python",
      "args": [
        "<YOUR_PATH_TO_MCP>/src/server.py"
      ]
    }
  }
}

The configuration on this page does not set TWELVE_DATA_API_KEY; the project's instructions put it in a .env file. Check which folder the server reads that file from before you rely on it.

Twelve Data