LangGraph Agent Review

Simple repository to review and test basic concepts of LangGraph agents.

Setup

1. Create Conda Environment

The recommended way to create the Conda environment is using the provided environment.yml file:

conda env create -f environment.yml
conda activate agents

This ensures you have the correct Python version and Poetry installed within the environment.

2. Install Dependencies

Once the conda environment is active and has Poetry, install the project dependencies:

poetry install

Running the Agent

LangGraph Development Server

You can run the agent using the LangGraph development server:

langgraph dev

FastAPI Application

Alternatively, run the FastAPI application which exposes the agent via an API endpoint:

fastapi dev app/api.py