Skip to content

feat: add quant-scanner + a-share-scanner — dual-market quantitative stock screening - #25

Open
MerkyorLynn wants to merge 2 commits into
badlogic:mainfrom
MerkyorLynn:feat/quant-scanner
Open

feat: add quant-scanner + a-share-scanner — dual-market quantitative stock screening#25
MerkyorLynn wants to merge 2 commits into
badlogic:mainfrom
MerkyorLynn:feat/quant-scanner

Conversation

@MerkyorLynn

@MerkyorLynn MerkyorLynn commented Apr 15, 2026

Copy link
Copy Markdown

Summary

Two model-driven quantitative stock scanning skills covering both US and Chinese markets:

  • quant-scanner: US equities (~100 large/mid-cap, Finnhub API, free tier)
  • a-share-scanner: Chinese A-shares (5000+ stocks, Tushare Pro, free tier)

How They Work

Both skills use the same architecture:

  1. Fetch real-time quotes + fundamental metrics from free API
  2. Detect market regime (volatility, momentum, breadth/money flow)
  3. Score each stock on adaptive multi-factor model (weights adjust to regime)
  4. Output ranked lists + HTML report + JSON

quant-scanner (US)

Factor Weight Adapts to
Momentum (5D/13W/MTD + S&P relative) 30% Bull x1.3 / Bear x0.7
Value (P/E, fwd P/E, PEG, div yield) 20% Bear x1.3 / Bull x0.8
Quality (ROE, margin, growth) 25% High-vol x1.2
Analyst (consensus ratio) 10% Stable
Risk (beta, vol, leverage) 15% High-vol x1.4
  • Zero external dependencies (Python stdlib only)
  • User provides free Finnhub API key

a-share-scanner (A-shares)

Factor Weight Adapts to
Momentum (daily + 5D + 20D) 25% Trending x1.3 / Declining x0.7
Volume (ratio + turnover + amount) 20% Stable
Valuation (PE/PB) 15% Declining x1.3
Growth (multi-period momentum) 20% High-vol x1.2
Price-Volume Confirmation 10% Stable
Risk (limit-up chase, excess vol) 10% High-vol x1.4
  • Requires: tushare + pandas
  • User provides free Tushare Pro token

Test plan

  • US: 91/93 stocks fetched successfully, HTML/JSON/text output verified
  • A-shares: 3080/5494 stocks scored (after ST/liquidity filter), HTML output verified
  • Both work with Python 3.9+
  • Graceful error handling for API rate limits

🤖 Generated with Claude Code

MerkyorLynn and others added 2 commits April 15, 2026 16:17
Model-driven quantitative stock scanner for US equities:
- Scans ~100 large/mid-cap stocks via Finnhub API (free tier)
- 5-factor adaptive scoring: momentum, value, quality, analyst, risk
- Market regime detection: auto-adjusts factor weights based on
  volatility, momentum, and breadth conditions
- Outputs HTML report, JSON, and LLM-readable text
- Zero external dependencies (Python stdlib only)
- User provides their own free Finnhub API key

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
…ning

Full A-share market scanner via Tushare Pro (free tier):
- Scans 5000+ stocks, filters to ~3000 after ST/liquidity removal
- 6-factor adaptive scoring: momentum, volume, valuation, growth, confirmation, risk
- Market regime detection with auto-adjusted factor weights
- Outputs HTML + JSON + LLM-readable text
- Requires: tushare + pandas (pip install)
- User provides free Tushare Pro token

Companion to quant-scanner (US stocks) for dual-market coverage.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
@MerkyorLynn MerkyorLynn changed the title feat: add quant-scanner — model-driven US stock quantitative screening feat: add quant-scanner + a-share-scanner — dual-market quantitative stock screening Apr 15, 2026
@burdiyan

Copy link
Copy Markdown

🤦‍♂️

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants