btse-mcp
An MCP server for the BTSE Futures API that allows AI agents to query market data, manage positions, and execute trades using natural language. It supports both live and testnet accounts with features for order management, leverage adjustment, and wallet balance tracking.
README
btse-mcp
MCP server for the BTSE Futures API. Enables AI agents (Claude Desktop, Cursor, LangChain) to query market data, manage positions, and place orders on BTSE via natural language.
Prerequisites
- Python 3.11 or higher — check with
python --version - pip — check with
pip --version - A BTSE account (testnet or live)
Multiple Python versions (Anaconda etc): use the full path explicitly:
/usr/local/bin/python3.14 -m pip install -e . /Library/Frameworks/Python.framework/Versions/3.14/bin/btse-mcp --helpUse that same full path in the Claude Desktop config (Step 4).
Step 1 — Get API keys from BTSE
Testnet (recommended first)
- Register at https://testnet.btse.io
- Go to Account → API tab → New API
- Save the API Key and Passphrase — the passphrase is shown only once and is your
api_secret - Set permissions: Read + Trading (add Transfer if needed)
Live
Same steps at https://btse.com
Step 2 — Install
# Clone the repo
git clone https://github.com/xbotlive/btse-mcp.git
cd btse-mcp
# Install — adds `btse-mcp` command to your PATH
pip install -e .
# Verify
btse-mcp --help
Virtualenv users: if
btse-mcpis not found after install, use the full path:
- macOS/Linux:
~/.venv/bin/btse-mcp- Windows:
~\.venv\Scripts\btse-mcp.exeYou will need this full path in the Claude Desktop config (Step 4).
Step 3 — Configure accounts
# Add a testnet account
btse-mcp config --account-id testnet
# Prompts:
# API Key → paste your API key
# API Secret → paste your passphrase (input is hidden)
# Use testnet? [y/N] → y
# Verify the connection — should print BTC-PERP last price
btse-mcp test testnet
# Add your live account when ready
btse-mcp config --account-id main
# Same prompts — answer 'n' to testnet
# See all configured accounts
btse-mcp list
Credentials are stored encrypted at ~/.config/btse-mcp/accounts.enc.
Unified Futures Wallet: If your BTSE account has been upgraded to the Unified Futures Wallet (all accounts from late 2024 onwards), account endpoints automatically use the v2.2 API. No action needed.
Step 4 — Connect to Claude Desktop
Find the config file for your OS:
| OS | Path |
|---|---|
| macOS | ~/Library/Application Support/Claude/claude_desktop_config.json |
| Windows | %APPDATA%\Claude\claude_desktop_config.json |
| Linux | ~/.config/Claude/claude_desktop_config.json |
Open the file (create it if it doesn't exist) and add:
{
"mcpServers": {
"btse": {
"command": "btse-mcp",
"args": ["start"]
}
}
}
Virtualenv users: replace
"btse-mcp"with the full path, e.g.:"command": "/Users/yourname/.venv/bin/btse-mcp"
Restart Claude Desktop. Open a new chat — you should see a tools icon (🔧) in the input bar.
Test it:
"What is the BTC-PERP mark price on BTSE using account testnet?"
Step 5 — Connect to Cursor
Open Cursor → Settings → MCP → Add Server and enter:
{
"name": "btse",
"command": "btse-mcp",
"args": ["start"]
}
Then use natural language in Cursor chat:
"Show my open BTSE positions" "Place a limit buy on BTC-PERP at 60000 size 1 using account testnet"
Alternative: run without installing
# From the repo root, no pip install required
python -m btse_mcp start
Tool list
| Tool | Description |
|---|---|
btse_get_market_summary |
Market summary for one or all symbols |
btse_get_price |
Mark / index / last price |
btse_get_orderbook |
L2 orderbook snapshot |
btse_get_trades |
Recent public trade fills |
btse_get_ohlcv |
OHLCV candlestick data |
btse_get_funding_history |
Historical funding rates |
btse_get_wallet_balance |
Futures wallet balance |
btse_get_positions |
Open positions |
btse_get_account_fees |
Maker / taker fee rates |
btse_get_leverage |
Current leverage for a market |
btse_create_order |
Place LIMIT / MARKET / OCO order (supports TP/SL) |
btse_cancel_order |
Cancel by order ID, or cancel all for a symbol |
btse_get_open_orders |
List open orders |
btse_get_order |
Single order detail |
btse_get_trade_history |
User trade history |
btse_amend_order |
Amend price / size / trigger price |
btse_close_position |
Close position at market or limit |
btse_set_leverage |
Set leverage (isolated or cross) |
btse_get_risk_limit |
Get risk limit tier |
All tools accept an optional account_id parameter (defaults to "default").
Pass "account_id": "testnet" to route to your testnet account.
Multi-account usage
btse-mcp list # list all configured accounts
btse-mcp test main # test a specific account
In prompts, specify the account explicitly:
"Using account testnet, show my BTC-PERP position"
Symbol naming
Use new-style perpetual names: BTC-PERP, ETH-PERP, SOL-PERP, etc.
Auth
BTSE uses HMAC-SHA384. The signature is:
HMAC-SHA384(api_secret, url_path + nonce + request_body)
Sent via headers: request-api, request-nonce, request-sign.
See docs/integration.md for full details and worked examples.
Running tests
pip install pytest
pytest -v
Auth signature tests run against the worked examples in the BTSE docs — no live API connection needed.
Troubleshooting
| Problem | Fix |
|---|---|
btse-mcp: command not found |
Use full path or activate your virtualenv |
401 Unauthorized |
Check API key and secret are copied correctly |
Connection failed |
Confirm testnet flag matches the account you created on |
| Tools icon missing in Claude Desktop | Check JSON syntax in config file, restart Claude Desktop |
ModuleNotFoundError: mcp |
Run pip install -e . again from the repo root |
33000001: Unsupported API |
Your account uses the Unified Futures Wallet — the server auto-retries on v2.2, restart Claude Desktop |
btse-mcp test works but account tools fail |
Restart Claude Desktop after any config or code change |
Disclaimer
Futures trading involves significant risk of loss. Always test on testnet before using live credentials.
Never commit API keys to version control — they are stored encrypted locally and excluded via .gitignore.
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.