MCP.so
ログイン

Freqtrade MCP Server

@kukapay

基本情報

カテゴリ

その他

ライセンス

MIT

トランスポート

stdio

作者

kukapay

Freqtrade MCP Server について

An MCP server that integrates with the Freqtrade cryptocurrency trading bot.

Freqtrade MCP Server の設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "freqtrade-mcp": {
      "command": "uv",
      "args": [
        "--directory",
        "/your/path/to/freqtrade-mcp",
        "run",
        "__main__.py"
      ],
      "env": {
        "FREQTRADE_API_URL": "",
        "FREQTRADE_USERNAME": "",
        "FREQTRADE_PASSWORD": ""
      }
    }
  }
}

Freqtrade MCP Server のツール

17

Fetch OHLCV data for a pair

Get open trade status

Get profit summary

Get account balance

Get performance metrics

Get whitelist of pairs

Get blacklist of pairs

Get trade history

Get bot configuration

Get trade locks

Place a buy/sell trade

Start the bot

Stop the bot

Reload bot configuration

Add pair to blacklist

Remove pair from blacklist

Delete a trade lock

概要

What is Freqtrade-MCP?

Freqtrade-MCP is an MCP (Model Context Protocol) server that integrates with the Freqtrade cryptocurrency trading bot via its REST API. It enables AI agents to interact with and operate Freqtrade for automated trading.

How to use Freqtrade-MCP?

Clone the repository, install dependencies (freqtrade-client and mcp[cli]), then configure your MCP client with the server’s path and environment variables (FREQTRADE_API_URL, FREQTRADE_USERNAME, FREQTRADE_PASSWORD). The Freqtrade instance must have its REST API enabled (set api_server.enabled = true in its configuration).

Key features of Freqtrade-MCP

  • 17 MCP tools provide access to most Freqtrade API endpoints.
  • Fetch market data, profit, balance, performance, trades, and configuration.
  • Place buy/sell trades and manage trade locks.
  • Start, stop, and reload the bot remotely.
  • Manage whitelist and blacklist of trading pairs.
  • Simple setup with environment‑variable authentication.
  • Works with Python 3.13+ and MCP clients (e.g., Claude Desktop).

Use cases of Freqtrade-MCP

  • Ask an AI assistant for live trading status and performance summaries.
  • Automate retrieval of market data (candlesticks) for analysis.
  • Allow natural‑language commands to place trades or adjust bot settings.
  • Integrate Freqtrade with conversational interfaces for hands‑off operation.
  • Programmatically add or remove pairs from blacklists via AI tools.

よくある質問

1

What tools does Freqtrade-MCP expose?

It exposes 17 tools: `fetch_market_data`, `fetch_bot_status`, `fetch_profit`, `fetch_balance`, `fetch_performance`, `fetch_whitelist`, `fetch_blacklist`, `fetch_trades`, `fetch_config`, `fetch_locks`, `place_trade`, `start_bot`, `stop_bot`, `reload_config`, `add_blacklist`, `delete_blacklist`, and `delete_lock`.

2

What are the prerequisites to run Freqtrade-MCP?

Python 3.13+, a running Freqtrade instance with the REST API enabled, and Git to clone the repository.

3

How is authentication handled?

Authentication is passed via environment variables: `FREQTRADE_USERNAME` and `FREQTRADE_PASSWORD` (matching the credentials in Freqtrade’s `api_server` configuration).

4

Can I use Freqtrade-MCP without a local Freqtrade instance?

No. The server requires a running, network‑accessible Freqtrade instance with its REST API enabled.

5

Is there any documentation for the Freqtrade REST API?

Yes. Freqtrade’s official documentation at `https://www.freqtrade.io/en/stable/rest-api/` explains how to enable and use the API that this MCP server wraps.

コメント

「その他」の他のコンテンツ