<2-3 sentences summarising what the project addresses and solves, the target audience, and the benefit.> TradeCat is an open-source, "toy-level" quantitative trading data platform designed for comprehensive market analysis and monitoring. It aims to provide users with tools to analyze, trade, and monitor diverse financial markets using a wide array of data sources and analytical methods. The platform targets individual traders, data analysts, and researchers seeking a unified solution for financial market intelligence.

How It Works

TradeCat employs a microservices architecture, integrating data from over 100 crypto exchanges (via CCXT, Cryptofeed), A-shares, US/global stocks, and macroeconomics. Data is stored in TimescaleDB and SQLite. Core functionalities include calculating 38 technical indicators using Python, pandas, and TA-Lib, performing AI-driven analysis (Wyckoff methodology) with support for multiple LLMs, and delivering insights via a Telegram bot. An optional order-service provides market-making capabilities.

Quick Start & Requirements

Installation is recommended via an AI-assisted prompt for tools like Claude or ChatGPT. Manual setup requires Python 3.10+ (3.12 recommended), PostgreSQL 16+ with TimescaleDB, TA-Lib 0.4+, and SQLite 3.x. Key steps involve cloning the repository, installing system dependencies (including TA-Lib compilation), initializing services with ./scripts/init.sh, configuring the .env file (crucially, deciding on and synchronizing TimescaleDB ports 5433/5434), and starting core services with ./scripts/start.sh start. Pre-built historical datasets are available on HuggingFace to bypass lengthy backfilling. WSL2 users need specific .wslconfig settings. Links to AI installation prompts and WSL2 setup guides are provided.

Highlighted Details

• Multi-Market Data Aggregation: Ingests data from 100+ crypto exchanges, A-shares, US/global stocks, and macro sources.
• Extensive Technical Indicators: Implements 38 indicators across trend, momentum, volatility, and volume, plus 61+ candlestick patterns and price patterns.
• AI-Powered Market Analysis: Integrates Wyckoff methodology with support for major LLMs (OpenAI, Gemini, Claude, DeepSeek) for deep market structure analysis.
• Real-time Telegram Bot: Offers 20+ ranking cards, custom signal pushes (breakouts, anomalies), interactive queries, and AI-driven market insights.
• Large Historical Datasets: Provides downloadable, pre-compressed historical K-line and futures metric datasets for rapid setup.
• High-Priority Symbol Identification: Automatically identifies ~130-150 key symbols based on volume, volatility, open interest, and sentiment metrics.

Maintenance & Community

The project is primarily maintained by tukuaiai. Community engagement is facilitated through Telegram channels (tradecat_ai_channel, glue_coding), Discord, and Twitter/X (123olp). GitHub Actions are configured for CI, though they perform only basic sampling checks.

Licensing & Compatibility

TradeCat is released under the permissive MIT License, allowing for broad use, modification, and distribution, including in commercial and closed-source applications.

Limitations & Caveats

The project is described as "toy-level," suggesting it may not be production-ready for high-stakes trading without further validation. A significant setup hurdle is the dual TimescaleDB port configuration (5433 vs. 5434), requiring careful synchronization across multiple scripts and configuration files. Some services are marked as "preview," indicating potential instability or incomplete features. The AI-driven installation method may not be universally applicable. TA-Lib installation can be complex.