How to use daily_stock_analysis: LLM-powered multi-market stock analysis and automated delivery

ZhuLinsen/daily_stock_analysis is an LLM-driven multi-market stock analysis system. It combines market data, news, market reviews, decision dashboards, and automated delivery. It is useful for personal research and automated information organization.

Project repository:

https://github.com/ZhuLinsen/daily_stock_analysis

Online preview:

https://dsa.zhulinsen.tech

Note: tools like this can only help organize information. They are not investment advice. Real trading decisions require your own risk judgment.

Run Locally

The README gives a clear local workflow:

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11


# Clone the project
git clone https://github.com/ZhuLinsen/daily_stock_analysis.git && cd daily_stock_analysis

# Install dependencies
pip install -r requirements.txt

# Configure environment variables
cp .env.example .env && vim .env

# Run analysis
python main.py

The key is .env. Fill in the configuration according to your own data sources, LLM service, and delivery channels. Do not rush into scheduled jobs. Run it locally once first.

Common Commands

Debug mode:

1

python main.py --debug

Dry run without actually sending notifications:

1

python main.py --dry-run

Specify stocks:

1

python main.py --stocks 600519,hk00700,AAPL

Generate a market review:

1

python main.py --market-review

Scheduled execution:

1

python main.py --schedule

Start only the service:

1

python main.py --serve-only

Web UI

If you want to view the dashboard in a browser, run:

1

python main.py --webui

Or start only the Web UI:

1

python main.py --webui-only

GitHub Actions Workflow

This project is also suitable for scheduled execution with GitHub Actions:

Fork the repository.
Configure API keys and delivery parameters in repository Secrets.
Enable Actions.
Run the workflow manually once.
After confirming output and delivery, let the scheduled job take over.

Usage Suggestions

A steadier approach is to test with a small stock list first, such as 600519,hk00700,AAPL. After confirming market data, news, LLM summaries, and delivery all work, expand the stock pool. The biggest risk with automation is “it looks like it ran, but the data source failed or the notification was wrong”, so it is worth periodically checking the raw data and conclusions.