daily_stock_analysis

LLM-driven stock analysis system. GitHub: ZhuLinsen/daily_stock_analysis

Overview

• Created: January 2026
• Stats: 34K stars, 33K forks (unusual ~1:1 ratio suggests artificial inflation)
• Deployment: GitHub Actions free tier, zero server cost
• Markets: A-shares, Hong Kong stocks, US stocks, cryptocurrency

Architecture

• 7 market data sources: TickFlow, AkShare, Tushare, baostock, efinance, pytdx, ccxt
• 7 news sources: Anspire, SerpAPI, Tavily, Jina AI, SearxNG, Google Custom Search, DuckDuckGo
• Social sentiment: Reddit/X/Twitter/Polymarket
• 6 push channels: WeCom, Feishu, DingTalk, Telegram, Discord, email

Output

Structured dashboard with: core conclusion, composite score (0-100), buy/sell price ranges, risk alerts, action checklists.

Caveats

“Zero cost” claim is misleading — data sources (SerpAPI, Tavily, Polymarket) require paid API keys. Free-only setup loses news aggregation and sentiment analysis.