Trading Assistant with LangChain, LangGraph & OpenBB API tools

An intelligent trading assistant powered by LangChain & LangGraph that integrates with OpenBB and Yahoo Finance for real-time market data. Supports multiple LLM providers: OpenAI, Google Gemini, Anthropic Claude, Perplexity, OpenRouter, and Ollama.

Features

• 🤖 Multi-LLM Support: Choose between OpenAI, Google Gemini, Anthropic Claude, Perplexity, OpenRouter, or Ollama
• � Multi-Agent System: Separate data retrieval from analysis for better performance and clarity
• 📊 Real-time Market Data: Access stock quotes, historical data, news, and company profiles via OpenBB Tools (using Yahoo FInance & OECD as two main sources)
• 📈 Economic Indicators: Fetch GDP, CPI, and other economic data
• 💬 Conversational Interface: Natural language queries for financial data
• 🌟 Commodity Mapping: Automatic symbol resolution (Gold→XAU, Silver→XAG, etc.)

Installation

1. Clone the Repository

In order to install the application, first make sure you have git, conda installed.

Then, clone the source code from GitHub onto your local machine and navigate into the Trading-Assistant directory. Finally, use the provided environment.yaml file to create the conda environment.

git clone https://github.com/samarthiith/Trading-Assistant
cd Trading-Assistant

2. Create Conda Environment

conda env create -f envs/environment.yaml
conda activate trading

3. Set Up Environment Variables

Copy the example environment file and add your API keys:

cp credentials/.env.example credentials/.env

Edit .env and add your API keys:

• OpenAI: https://platform.openai.com/api-keys
• Google Gemini: https://makersuite.google.com/app/apikey
• Anthropic Claude: https://platform.claude.com/docs/en/api/admin/api_keys/retrieve
• Perplexity: https://www.perplexity.ai/settings/api
• OpenRouter: https://openrouter.ai/keys

Note: Ollama runs locally and doesn't require an API key.

4. Set Up Model Config

Select the LLM model that you would like to us in the config.yaml.

Usage

1. Basic Command-line (CLI) Usage

Run the agent:

python agent_system.py

You'll be prompted to choose between:

• Two-Agent System: Separate data retrieval and analysis

Example interaction with Two-Agent System:

🤖 Initializing Agent System with gemini...
✅ Agent System ready!
   📡 Data Agent: Fetches market data
   📊 Analysis Agent: Provides insights

You: What's the gold price trend?
[Agent processes with two-phase approach: data → analysis]

You: Get me the latest news about Tesla
[Agent retrieves TSLA news then analyzes sentiment]

2. Running the Streamlit Dashboard

To run the Streamlit dashboard:

streamlit run dashboard.py

The dashboard will open in your browser at http://localhost:8501. The app will enable you to interact in the chatbox.

In the sidebar, you can switch between:

• Two-Agent System: Separate data fetching and analysis

Agent Modes

Modern architecture separating concerns:

graph TD
    A[User Query] --> B[Orchestrator]
    B --> C[Data Agent]
    C -->|retrieves data via tools| D[Analysis Agent]
    D -->|analyzes data| E[Combined Response]

Loading

Benefits:

• Clear separation between data and analysis
• Specialized prompts for each task
• Better debugging and transparency
• More accurate commodity symbol mapping

See MULTI-AGENT_SYSTEM.md for detailed documentation.

License

AGPLv3 License