This sample application does stock analysis by conducting news search, fundamental review and technical analysis. This app is built on Single Supervisor with multiple agents architecture as described here. The multiple agents used in this app uses OpenAI LLM Model and uses TavilySearch tool to search for news. Additionally it uses Yahoo Finance for fundamental and technical analysis.
The following API keys are required for this app
- OPENAI_API_KEY
- Python 3.8 or higher
The app uses streamlit to create a web page so that users can interact with the app using browser. Once the app is launched, it spits out the url to access the app. Open a browser window and put the url to access the UI of this app.
The app uses traceloop to send telemetry data. If you want to capture those telemetry data please add the endpoints in .env file.
To get tracing data in Dynatrace, provide these keys:
- DYNATRACE_EXPORTER_OTLP_ENDPOINT (e.g .live.dynatrace.com/api/v2/otlp)
- DYNATRACE_API_TOKEN
Copy .env.template to .env Update .env file with all the required keys
python -m venv financialagent
source financialagent/bin/activate
pip install -r requirements.txt
streamlit run main.py
./run.sh
- Analyze APPL stock
- Give me latest news on Nvidia
- Conduct fundamental analysis on DIS