GitHub radar
Free LLM Stock Analysis: Daily Push to Telegram
An LLM-powered stock analysis system that reads news, calculates indicators, and sends a daily decision dashboard to Telegram — free to run via GitHub Actions.
This is an LLM-powered system for daily stock analysis: each working day it reads news and calculates technical indicators on its own, then sends a decision report for each ticker to Telegram, WeChat, Discord, Slack, or email. The report contains a score, trend direction, recommended entry and exit points, risks, upcoming catalysts, and an action checklist. It supports stocks from the A-share market, Hong Kong, the United States, Japan, South Korea, and Taiwan. The system works with any OpenAI-compatible model, including DeepSeek, Claude, Gemini, GPT, and local Ollama models. The recommended way to run it is through GitHub Actions, at zero cost and without a server, requiring only an AI model key and notification channel credentials. There is also a web interface with report history, manual runs, and a chat mode for asking the agent about a specific stock. The main README is in Chinese, but a full English translation is available. The project is MIT-licensed.
Why a vibe-coder should care
The system saves the time that goes into daily portfolio monitoring: instead of reading news and checking charts yourself, you receive a ready analysis in your messenger. You set it up once through GitHub Actions and it works by itself, with no server and no monthly costs.
How to install
Copy this and send it to your agent — Claude Code, Codex, any of them:
Deploy the stock analysis system from https://github.com/ZhuLinsen/daily_stock_analysis — configure a ticker list (e.g. AAPL, NVDA), connect my AI model key, and set up Telegram notifications via GitHub Actions. Ask me if you need API keys.
Runs on GitHub Actions at no cost — you only need a browser to set it up and an AI model key. For local runs: a regular laptop with 8 GB RAM or more.
Open on GitHub▌ More finds