financial chat

financial-chat

A financial chat application powered by LangChain, LangGraph, OpenBB, Claude 3 Opus, and Streamlit.

Blog Articles

If you're curious about the journey of building this project, check out these blog articles:

• Building an Agentic Stock Analysis Tool with LangChain, OpenBB, and Claude 3 Opus
• Expanding the AI Stock Analysis Agent with Fundamental and Technical Tools
• Improving the Edge: Trendlines, Sectors, and Sentiment
• Unlocking Alpha: Harnessing Relative Strength for AI-Driven Stock Selection
• Seeing the Bigger Picture: Adding Chart-Based Technical Analysis
• Empowering the AI Stock Analysis Agent with Universe Scanning
• Risk Management in AI Stock Trading: A Key to Success
• Converting the AI Stock AgentExecutor to LangGraph
• Unleashing the Power of Multiple Agents with LangGraph
• Deploying the AI Stock Analysis Agent on AWS with Copilot

Features

• Fetches financial data using OpenBB
• Generates technical analysis summaries using AI
• Provides stock price history, quantitative stats, and more
• Calculates relative strength for stocks
• Sentiment analysis on news articles
• Universe scanning using FinViz filters
• Risk management techniques using technically-derived stops and R Multiples
• Interactive Streamlit UI for chat-based interaction
• Multiple Agent Workflows using LangGraph
• Deployment to AWS with the Copilot CLI

Installation

1. Install the required dependencies using Poetry:

poetry install

2. Set up the necessary environment variables. You can create an .env at the project root for these:

export OPENAI_API_KEY=<your-api-key>
export OPENBB_TOKEN=<your-openbb-token>
export TIINGO_API_KEY=<your-tiingo-api-key>
export IMGUR_CLIENT_ID=<your-imgur-client-id>
export IMGUR_CLIENT_SECRET=<your-imgur-client-secret>
export FMP_API_KEY=<fmp-api-key>
export INTRINIO_API_KEY=<intrinio-api-key>

Usage

Streamlit UI

Run the Streamlit app:

streamlit run app/ui.py

FastAPI Server

Start the FastAPI server:

uvicorn app.server:app --host 0.0.0.0 --port 8080

You can view the Swagger Docs and test out the Playground, courtesy of LangServe.

Docker

Build the Docker image:

docker build -t financial-chat .

Run the Docker container:

docker run -p 8080:8080 --env-file .env financial-chat

Project Structure

• app/: Main application code
  • chains/: LangChain agent and prompts
  • features/: Feature-specific code (technical analysis, charting)
  • tools/: Custom tools for data retrieval and analysis
  • ui.py: Streamlit UI
  • server.py: FastAPI server
• Dockerfile: Dockerfile for building the application
• pyproject.toml: Project dependencies and configuration
• README.md: Project documentation

Flowchart