terancejiang

财报排雷工具 - A-share financial report fraud/risk detection based on Tang Chao's methodology. Claude Code Skill with 28-rule checklist.

17
6
100% credibility
Found Apr 12, 2026 at 17 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

This project creates an AI chat command that automates 30 financial risk checks from a bestselling book to screen Chinese stocks for potential dangers.

How It Works

1
📖 Discover the Stock Safety Checker

You stumble upon this clever tool inspired by a popular book that helps everyday investors spot hidden risks in company financial reports, like clearing mines in a game.

2
🛠️ Run the Simple Setup Helper

Download the project and launch the one-click setup that checks your computer and prepares the AI assistant in seconds.

3
🔗 Connect Free Stock Info Service

Sign up for a free stock data account, paste your private access code, and now the tool can pull real company numbers automatically.

4
🤖 Ask the AI to Check a Stock

In your friendly AI chat, just type a stock name like 'Guizhou Moutai' or code like '600519', and it springs to life.

5
🔍 See the Magic Analysis Unfold

The assistant grabs years of financial details, applies 30 book-based safety checks, and even peeks at the latest company report.

📈 Receive Your Clear Risk Report

Get a straightforward summary with risk level, key warnings, and investment insights, all saved neatly for you to review and share.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 17 to 17 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is financial-report-minesweeper?

This Python tool automates A-share financial report fraud detection using Tang Chao's methodology as a 28-rule checklist. In Claude Code, type /minesweeper 600519 (or company name like 贵州茅台), and it fetches 10 years of Tushare data plus annual PDFs to score risks across profit signals, cash flows, and non-financial flags, outputting a graded report to an organized folder. It solves manual Excel drudgery for spotting fakes in Chinese listings.

Why is it gaining traction?

Claude Code skill integration delivers instant reports without setup hassles, unlike raw Tushare scripts—perfect for quick A-share screens. The bash install.sh handles deps and token setup, making it easy to share a GitHub repo with someone or link a private GitHub repository in trading groups. Chao's rules (audit fails, cash mismatches) provide structured edge over vague scanners.

Who should use this?

A-share retail traders screening 100+ tickers weekly, or hobby quants automating due diligence before buys. Great for Discord crews sharing GitHub Copilot chats or codespaces on stock ideas—run on 000858 2023, share the report. Skip if you need enterprise compliance.

Verdict

Grab it for fast A-share GitHub workflows; 17 stars and 1.0% credibility score mean it's raw (no tests, solo dev), but excellent README and outputs make it forkable now. Maturity lags, yet punches hard for checklist-driven checks.

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.