V3 Quantitative Trading System
Paste the following prompt into your AI chat to install this skill:
Please install @user_4719b322/v3-quant-trading by following https://skillhub.cn/install/skillhub.md.
About this skill
Problem This Skill Addresses
V3 Quantitative Trading System is aimed at local workflows for A-share and U.S. equity quote analysis, scoring, order execution, and trade review. It is not just a quote viewer; it connects a FastAPI backend, React frontend, market data, broker interfaces, and trade memory into a debuggable quant pipeline. When a user needs to decide whether a ChiNext or STAR Market symbol fits a short-horizon trading setup, the system can pull real-time quotes, score the setup, produce buy/sell suggestions, and record the trade context after execution.
How It Works and Where the Limits Are
The workflow covers automatic stock selection and trading, manual market analysis, and agent self-learning. The manual path runs through symbol input, real-time quote lookup, a 10-dimension score, and trade recommendation generation. The scoring dimensions include price position, trend structure, bid-ask pressure, volatility, volume-price relationship, sector environment, contrarian signals, index resonance, intraday timing, and holding state, with the full logic in references/scoring-engine.md. Risk controls are split by board: ChiNext day-trading defaults include an 8% target, 3% stop-loss, 5% take-profit, and a 2.5% trailing-drawdown trigger; STAR Market uses a wider stop-loss, smaller per-position cap, and a higher take-profit target. Market data comes from Sina Finance, covering SSE, SZSE, ChiNext, STAR Market, and U.S. equities, with up to 50 symbols per batch and an approximately 30-second cache TTL, but it does not support Beijing Stock Exchange or HK markets. The system also abstracts broker access through SIM, XtQuant, and XTP, and stores trade experience for weight adaptation and market-state recognition. The practical boundary is local operations and strategy debugging: it requires Python, Node.js, and local configuration; sensitive files such as backend/.env must be protected; and live-trading integrations should be validated for broker compatibility, quote coverage, and risk parameters before use.
Use Cases
- Start the local backend and frontend, query a ChiNext symbol's live quote, and generate a buy/sell recommendation.
- Batch-fetch Sina quotes for multiple SSE/SZSE symbols and compare trade signals with the 10-dimension score.
- Run ChiNext day-trading in the SIM account, tracking 3% stop-loss and 2.5% trailing drawdown rules.
- Review executed trades, inspect agent insights, and adjust scoring weights and market-state judgments.
Best For
- Quant engineers who want to debug short-horizon A-share strategies and validate risk-control parameters locally
- Integration engineers wiring Sina quotes to SIM, XtQuant, or XTP broker interfaces
- ML engineers converting executed trades into memory and reviewing adaptive scoring weights
- Researchers inspecting ChiNext/STAR risk parameters and generating buy/sell recommendations
Related Skills
Model routing, persistent parameter management, self-check repair, and global default model control for XiaoYi Claw.
Install, update, and manage OpenClaw Skills through SkillHub, with automatic detection after installation.
Provides API endpoints for AI agents to post bottles and graffiti, browse the feed, and interact with likes and comments.
ReqPlan constrains agent development, debugging, and analysis workflows with a seven-stage state machine, checkpoints, quality audits, and local Harness artifacts.