dasein108/freqtrade_dev_mcp

本页由 AI 翻译，技术细节请以原文为准。查看原文

10 stars 4 forks 质量 80 趋势 16MIT

MCP server exposing Freqtrade backtesting, hyperopt and market-data tools to Claude and other LLM clients, plus a LangGraph agent that generates and optimizes trading strategies

概览

MCP server that exposes Freqtrade strategy development tools (data download, backtesting, hyperopt, result analysis) to LLM clients like Claude Desktop and Claude Code. It ships with a LangGraph agent that uses the same tools to generate, optimize and evaluate strategies on its own. Independent project, not affiliated with the Freqtrade team. For research and education. Backtest results do not predict live performance. Never trade real funds with a strategy you have not reviewed yourself. Full parameter reference: docs/tools-reference.md. - Python 3.11+ (required by freqtrade>=2025.7) - A Freqtrade installation with a user_data/ directory - freqtrade on PATH for backtest/hyperopt runs - For the agent: an API key for one of the supported LLM providers pip install -r requirements.txt also works and installs everything, agent included. The server reads ~/.config/freqtrade-mcp/config.

README

以下为原文 README 节选，完整内容请查看原文。

Freqtrade Dev MCP

Independent project, not affiliated with the Freqtrade team. For research and education. Backtest results do not predict live performance. Never trade real funds with a strategy you have not reviewed yourself.

What’s inside

Component	Path	Purpose
MCP server	`src/`	stdio MCP server with 12 Freqtrade tools
Strategy agent	`strategy_agent/`	LangGraph workflow: idea → code → hyperopt → backtest → analysis → rewrite
Examples	`examples/`	Claude Desktop / Claude Code configs, usage scripts
Docs	`docs/`	Tool reference, developer guide, agent docs
Tests	`tests/`	Unit and integration tests (pytest)

MCP tools

Group	Tool	What it does
Setup	`create_userdir`	Create a Freqtrade `user_data` directory
	`create_config`	Generate a config from a template (`default`, `conservative`, `aggressive`, `advanced`)
Strategy	`create_strategy`	Generate a strategy from a template (`basic`, `trend`, `mean_reversion`, `scalping`, `advanced`)
	`create_strategy_wireframe`	Generate a minimal skeleton for an LLM to fill in
Data & runs	`download_candles`	Download OHLCV data; accepts natural-language ranges (`"last 3 months"`) and `top15` by market cap (CoinGecko)
	`backtest_strategy`	Run a backtest, optionally exporting trades and signals
	`hyperopt_strategy`	Run hyperparameter optimization
Analysis	`extract_backtest_data`	Parse a backtest `.zip` into metrics, trades, per-pair / per-hour stats
	`extract_hyperopt_data`	Parse a `.fthypt` file into best params, parameter ranges, convergence
	`search_results`	Filter indexed results by profit, drawdown, win rate, trades, dates (SQLite index)
	`list_results` / `get_result`	Browse and fetch saved results

Full parameter reference: docs/tools-reference.md.

Requirements

Python 3.11+ (required by freqtrade>=2025.7)
A Freqtrade installation with a user_data/ directory
freqtrade on PATH for backtest/hyperopt runs
For the agent: an API key for one of the supported LLM providers

Installation

git clone https://github.com/dasein108/freqtrade_dev_mcp.git
cd freqtrade_dev_mcp

# MCP server only
uv pip install -e .

在 GitHub 查看完整 README

安装

This server does not publish a one-line install command.

Open the repository installation guide 新手？先看 mcpskill.org 教程 →

配置

{
  "mcpServers": {
    "freqtrade": {
      "command": "/absolute/path/to/python",
      "args": ["/absolute/path/to/freqtrade_dev_mcp/run_server.py"],
      "env": {
        "FREQTRADE_MCP_PATH": "/absolute/path/to/freqtrade"
      }
    }
  }
}