A Streamlit app for natural language querying of Saudi XBRL financial data using PandasAI and OpenRouter/Gemini.
- Query Saudi listed companies' financial data using natural language
- Multiple datasets: Analytics View, Filings, Financial Facts, Financial Ratios
- Interactive charts and data visualization
- Powered by PandasAI + Gemini 3 Flash
- Go to GitHub and create a new repository
- Name it something like
saudi-financial-chat - Make it public (required for Streamlit Cloud free tier)
Upload all files from this folder to your GitHub repository:
app.py- Main applicationrequirements.txt- Python dependencies.streamlit/config.toml- Streamlit configurationdata/folder with all parquet files
- Go to Streamlit Cloud
- Sign in with your GitHub account
- Click "New app"
- Select your repository and branch
- Set main file path to
app.py - Click "Deploy"
After deployment, add your OpenRouter API key:
- In Streamlit Cloud, go to your app settings
- Click "Secrets" in the left sidebar
- Add the following:
OPENROUTER_API_KEY = "sk-or-v1-your-actual-api-key"- Click "Save"
pip install -r requirements.txt
streamlit run app.py- Saudi Tadawul XBRL Financial Reports
- Pre-processed parquet files for fast loading