股海罗盘 - A股股票量化分析
A quantitative analysis tool with historically validated signals for A-share investment.
Install & Use
Copy this prompt and send it to your AI assistant (Claude / Cursor / TRAE / Codex / WorkBuddy etc.) to auto-install:
Help me install this AI Skill: 股海罗盘 - A股股票量化分析. It is used for: A quantitative analysis tool with historically validated signals for A-share investment. Full Skill content: https://321skill.com/skills/a-share-quant-analysis-15/raw/index.md Read that page and install it.
The prompt includes a link to the full Skill content. You can also view the full content.
Guhai Luopan is a self-learning quantitative analysis tool designed for the A-share market. It addresses three core pain points investors face in decision-making: unclear signal validity, fragmented information, and difficulty in retaining analysis results. By backtesting vast amounts of historical data, the tool attaches a historical match rate to each technical or capital flow signal, allowing users to intuitively understand the signal's past performance, thereby reducing blind following.
To use it, simply input a stock code or name. Guhai Luopan will automatically invoke its quantitative engine to backtest over 600,000 historical signals. It integrates data across 17 dimensions including market quotes, capital flows, technical indicators, financials, and research reports, generating an in-depth report containing 5 types of pattern charts (e.g., support/resistance levels, probability after rise/fall) and exclusive ETF four-quadrant linkage analysis. The entire process eliminates the need to switch between different software or websites.
This tool is well-suited for investors with some understanding of the A-share market who wish to leverage data to aid decision-making, particularly quantitative traders, data analysts, and financial professionals who need to provide clients with professional analysis reports. It transforms complex market information into verifiable, quantifiable conclusions.
It is recommended to use the tool as a supplementary reference for investment decisions, not as the sole basis. Initially, users can test it with a few familiar stocks, comparing the generated pattern charts with their own understanding to better grasp the tool's output logic. As the analysis is based on historical data, users should integrate it with the current macroeconomic environment for comprehensive judgment.
Key Features
The core differentiator is providing a visual 'Historical Match Rate' progress bar for each analysis signal, clearly informing users how many times the signal was accurate out of the past 100 occurrences, rather than simply presenting an untested indicator or conclusion. Additionally, it uniquely integrates ETF four-quadrant linkage analysis and cross-validation with research reports from all market brokerages, offering in-depth verification from both capital flow and market consensus perspectives.
Limitations
The tool primarily relies on historical data for pattern discovery and backtesting and is not suitable for predicting future unexpected black swan events or extreme policy-driven market conditions.
FAQ
How is the Historical Match Rate calculated?
It is based on the proportion of times the stock price reached the expected trend within a set period after similar historical signals appeared for that stock, presented intuitively with a progress bar.
What does the generated DOCX report include?
The report includes candlestick charts, a summary of 17-dimensional core data, analysis of 5 types of pattern charts, details of the Historical Match Rate, ETF fund linkage analysis, and a summary of brokerage viewpoints.
Installation guide for AI assistants
If your AI coding assistant (Claude Code, Cursor, TRAE etc.) can see this page, send it this message to auto-install:
Visit https://321skill.com/skills/a-share-quant-analysis-15/raw/index.md to read the original Skill definition (Markdown format) for 股海罗盘 - A股股票量化分析, and install it according to the instructions.
Raw Markdown URL for AI: /skills/a-share-quant-analysis-15/raw/index.md