The Finance Agent Team is a Streamlit-based application that leverages the deepseek-r1-distill-llama-70b model from Groq to provide users with advanced conversational capabilities. This app integrates two specialized agents:
- Web Agent: Searches the web for information using DuckDuckGo.
- Finance Agent: Provides financial data such as stock prices, analyst recommendations, company information, and news using YFinanceTools.
Both agents are backed by Groq's powerful LLM and are designed to deliver accurate and contextual insights.
- Web Search Integration: The Web Agent fetches real-time information from the web.
- Financial Data Retrieval: The Finance Agent delivers comprehensive insights into stocks and companies.
- Agent Team Collaboration: Combines the strengths of both agents to answer complex multi-domain queries.
- SQLite Storage: Saves conversation history for both agents using SQLite, ensuring context awareness.
- Streamlit UI: Clean and interactive user interface for seamless query submission and response visualization.
Ensure you have Python 3.8 or higher installed.
-
Clone the repository:
git clone https://github.com/your-username/finance-agent-team.git cd finance-agent-team -
Install the required Python libraries:
pip install phi streamlit groq
-
Set up the Groq API key:
export GROQ_API_KEY="your_groq_api_key_here"
-
Run the Application:
streamlit run app.py
-
Interface:
- Enter your query in the provided text box.
- Click Submit to receive a detailed response from the agent team.
- Role: Searches the web for real-time information.
- Tools: DuckDuckGo.
- Storage: Saves conversation history in an SQLite database (
agents.db).
- Role: Provides financial insights.
- Tools: YFinanceTools (stock prices, analyst recommendations, company information, news).
- Storage: Saves conversation history in an SQLite database (
agents.db).
- Combines the capabilities of the Web and Finance agents.
- Designed for multi-domain queries with tool call visibility enabled.
- User Input: "What is the current stock price of Tesla and the latest news?"
- Agent Team Response:
- Table displaying Tesla's stock price.
- Links to the latest news articles about Tesla.
- The app uses GroqChat for conversational AI.
- Ensure the Groq API key is correctly configured.
- SQLite database (
agents.db) is automatically created to store interaction history.
- Add support for additional tools (e.g., weather updates, news summarization).
- Implement user authentication for personalized experiences.
- Expand financial data coverage (e.g., cryptocurrency, global markets).
This project is licensed under the MIT License.
For queries or contributions, feel free to reach out:
- Author: Siddharth Kharche
- Email: [email protected]
- GitHub: siddharth-Kharche