This project provides a full-stack application for stock market analysis, leveraging LLMs, LangChain, and LangGraph to retrieve and visualize stock data and news. It targets users seeking AI-driven insights into financial markets, offering features like stock performance charting and attribute-specific data retrieval.

How It Works

The application employs agentic Retrieval-Augmented Generation (RAG) workflows. News data is scraped asynchronously, stored in MongoDB, and synchronized with ChromaDB for semantic search. Financial data is scraped and stored in PostgreSQL. LangGraph orchestrates three main RAG graphs: one for news (retrieving from ChromaDB or web search), one for stock data (generating and executing SQL queries), and one for generating stock charts from SQL data.

Quick Start & Requirements

• Install dependencies via pip install -r requirements.txt.
• Run tests with pytest.
• Requires Python 3.x.
• Integrates with LangSmith for observability.
• Official API documentation is available via openapi.json.

Highlighted Details

• Agentic RAG workflows for news and financial data analysis.
• LangGraph for orchestrating complex, stateful LLM agent interactions.
• Asynchronous data scraping for news (MongoDB) and financial data (PostgreSQL).
• Includes API endpoints for retrieving stock price statistics, charts, and news.
• Utilizes LangSmith for detailed LLM call tracing and debugging.

Maintenance & Community

• The project is maintained by vinay-gatech.
• LangSmith integration suggests a focus on observability and debugging LLM workflows.

Licensing & Compatibility

• Licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
• This license restricts commercial use and requires derivative works to be shared under the same terms.

Limitations & Caveats

• The non-commercial license significantly limits its applicability for business or enterprise use.
• Specific LLM providers or API keys are not explicitly mentioned, implying potential configuration requirements.