This project demonstrates a powerful AI agent interface built with Agno and Nebius, featuring two specialized agents:
- A Web Agent for internet research and information gathering
- A Finance Agent for stock market analysis and financial data
- Interactive UI for interacting with AI agents
- Web search capabilities using DuckDuckGo
- Financial data analysis using YFinance
- Conversation history tracking
- Markdown formatting support
- SQLite storage for persistent conversations
- Python 3.8+
- Nebius API key
- Node.js and npm (for UI)
- Clone the repository:
git clone https://github.com/Arindam200/awesome-ai-apps.git
cd simple_ai_agent/agno_ui_agent- Install Python dependencies:
pip install -U openai duckduckgo-search yfinance sqlalchemy 'fastapi[standard]' agno- Set up environment variables:
Create a
.envfile in the project root and add your Nebius API key:
NEBIUS_API_KEY=your_api_key_here
- Install and run the Agent UI:
npx create-agent-ui@latest
cd agent-ui && npm run devOpen http://localhost:3000 to view the Agent UI.
- Start the Python backend:
python playground.py- Open http://localhost:3000 in your browser
- Select the http://localhost:7777 endpoint
- Choose between the Web Agent or Finance Agent
- Start interacting with your chosen agent
- Performs web searches
- Provides information with sources
- Maintains conversation history
- Retrieves stock prices
- Provides analyst recommendations
- Displays company information
- Shows relevant company news
- Presents data in formatted tables
The agents are configured in playground.py with the following features:
- Conversation history tracking (last 5 responses)
- Markdown formatting
- SQLite storage for persistence
- Automatic datetime inclusion
- Custom instructions for each agent
💡 Note: Make sure the serve_playground_app() points to the file containing your Playground app.