Sequoia-X

Product Introduction

1. Definition: Sequoia-X is an automated, open-source quantitative stock screening system specifically engineered for the Chinese A-share market. It is a Python-based technical analysis platform that leverages vectorized computation and an Object-Oriented Programming (OOP) architecture.
2. Core Value Proposition: It solves the problem of manually scanning hundreds of A-share stocks for complex technical patterns by automating the entire process. The system runs automated scans after market close and delivers actionable alerts directly to collaboration platforms like Feishu (Lark), saving significant time for retail investors and quantitative traders.

Main Features

1. Automated Post-Market Screening: The system is designed to run automatically after the A-share market closes. It executes multiple quantitative strategies in parallel, scanning the entire universe of ~5200 stocks without manual intervention.
2. Multi-Strategy Engine: Sequoia-X V2 includes a suite of six built-in, professionally-inspired trading strategies. These include the TurtleTrade breakout strategy, MaVolume moving average and volume breakout, HighTightFlag pattern recognition, LimitUpShakeout for pullback confirmation, UptrendLimitDown for trend reversal, and RpsBreakout for relative strength momentum.
3. Robust Data Pipeline with Baostock: The data layer uses the free, unlimited baostock API (which requires no registration) to pull historical and incremental daily K-line data. It employs post-adjusted (hfq) prices to prevent data distortion from stock splits and dividends, making it ideal for incremental storage and long-term backtesting.
4. Local SQLite Database & Incremental Updates: All market data is stored locally in a SQLite database (data/sequoia_v2.db), ensuring data ownership, portability, and circumventing anti-crawling restrictions from other financial data providers. The system uses 8-process parallelization for efficient incremental data updates, completing full-market sync in 2-3 minutes.
5. Integrated Feishu/Lark Notification: The system integrates seamlessly with Feishu (Lark) via webhooks to push formatted, actionable stock alerts directly to group chats or individuals, enabling real-time decision-making for traders and teams.
6. Dual Operational Modes: It supports two key modes: --backfill mode for initial historical data population (completing in ~12 minutes) and daily mode for incremental updates and strategy execution. This is managed through a clean command-line interface (CLI) via argparse.

Problems Solved

1. Pain Point: Manually screening thousands of A-share stocks for specific technical setups is incredibly time-consuming, error-prone, and inefficient for individual traders and small teams.
2. Target Audience: The primary users are retail investors and quantitative traders focused on the Chinese stock market. Secondary users include financial analysts and developer-traders who wish to customize or extend the open-source strategy framework.
3. Use Cases: A retail investor can automate their nightly stock screening routine. A quant trader can use it as a robust signal generation layer for a larger trading system. A trading team can receive consolidated alerts in a Feishu group for collaborative discussion and rapid execution.

Unique Advantages

1. Differentiation: Unlike many proprietary screening tools or platforms with high costs and API limits, Sequoia-X is free, open-source, and uses the unlimited baostock API. Its local SQLite storage provides full data control and privacy, unlike cloud-based SaaS solutions.
2. Key Innovation: The system's architecture is built on modern Python engineering standards (OOP, Pydantic for config, Rich for logging) and emphasizes vectorized computation for performance. The combination of a free, reliable data source (baostock) with a portable local database and automated multi-strategy execution in a single, cohesive package is its core technical innovation.

Frequently Asked Questions (FAQ)

1. Is Sequoia-X free to use for A-share stock screening? Yes, Sequoia-X is completely free and open-source (MIT licensed). It uses the free baostock API for data, requiring no registration or payment, making it a cost-effective solution for A-share quantitative analysis.
2. How does Sequoia-X get stock data and avoid being blocked? Sequoia-X pulls historical and daily K-line data from the baostock API, which is a free and official-friendly data source. It stores all data locally in a SQLite database, eliminating repeated calls to potentially restrictive sites like East Money and ensuring stable, reliable access.
3. What are the system requirements to run Sequoia-X? You need Python 3.10 or higher. The project recommends using the uv package manager for fast dependency installation. It can run on any machine capable of running Python, from a local PC to a cloud server for scheduled cron jobs.
4. Can I add my own trading strategies to Sequoia-X? Yes. The system is built with an extensible OOP architecture. You can create new strategies by subclassing the base strategy class in sequoia_x/strategy/base.py and integrating them into the main execution pipeline.
5. How do I set up Sequoia-X to run automatically after the market closes? You can use a cron job (on Linux/macOS) or a scheduled task (on Windows) to execute python main.py in the project directory after 3:00 PM China time (post-market). The repository README provides an example cron command for automated daily execution.