freqtrade-mcp

freqtrade-mcp

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

93
Stars
25
Forks
0
Releases

Overview

Freqtrade-MCP is an MCP server that bridges the Freqtrade cryptocurrency trading bot's REST API with the Model Context Protocol (MCP), enabling AI agents to interact with and govern automated trading operations. It relies on the freqtrade-client package and an MCP CLI, with an optional uv runner for execution. The setup requires Python 3.13+, a running Freqtrade instance with REST API enabled, and configuration for FREQTRADE_API_URL and credentials. The server exposes a suite of MCP tools that mirror common Freqtrade endpoints, including data retrieval (fetch_market_data, fetch_bot_status, fetch_profit, fetch_balance, fetch_performance, fetch_whitelist, fetch_blacklist, fetch_trades, fetch_config, fetch_locks) and trading actions (place_trade, add_blacklist, delete_blacklist, delete_lock). It also includes bot control (start_bot, stop_bot) and configuration reload (reload_config). Security is handled via api_server settings (JWT key, credentials, CORS, ws_token). Example prompts illustrate usage, and the repository provides integration steps and a sample REST API configuration. The MIT license applies.

Details

Owner
kukapay
Language
Python
License
MIT License
Updated
2025-12-07

Features

MCP surface mapped to Freqtrade REST API

Exposes a broad set of MCP tools that directly map to Freqtrade REST API endpoints, enabling AI agents to access on-data and bot state.

Comprehensive MCP toolset

Includes tools for market data, bot status, profit, balance, performance, whitelist/blacklist, trades, config, locks, and more.

Trading actions and order placement

Provide place_trade and management tools such as add_blacklist, delete_blacklist, and delete_lock for operational control.

Bot lifecycle and config management

Control bot lifecycle with start_bot, stop_bot, and reload_config to keep trading logic up to date.

Access control and API security

Configurable API server settings (JWT, credentials, CORS, etc.) to secure MCP access.

Fast installation and client integration

Install via freqtrade-client and mcp[cli], with optional uv runner for quick deployment.

Whitelist/Blacklist and lock management

Manage trading pairs and trade locks through dedicated MCP tools for risk control.

Audience

AI agentsEnable AI agents to fetch data, place trades, and manage the Freqtrade bot through MCP endpoints.
DevelopersBuild integrations and tooling around Freqtrade by consuming the MCP toolset.
Quant/tradersMonitor bot state, performance, balance, and trades to automate workflows.

Tags

FreqtradeMCPAI agentREST APITrading botCryptoPythonCLI