Python RAG Agent using LangGraph and Auth0 FGA

This example demonstrates how to combine LangChain with robust authorization controls for RAG workflows. Using Auth0 FGA, it ensures that users can only access documents they are authorized to view. The example retrieves relevant documents, enforces access permissions, and generates responses based only on authorized data, maintaining strict data security and preventing unauthorized access.

Getting Started

Prerequisites

• An Auth0 FGA account, you can create one here.
• An OpenAI account and API key create one here.

Setup

1. Create a .env file using the format below:

    # OpenAI
    OPENAI_API_KEY=xx-xxxx-xxxxxxxxxxxxxxxxxxxxxxxxxxx
   
    # Auth0 FGA
    FGA_STORE_ID=xxxxxxxxxxxxxxxxxxxxxxxxxxx
    FGA_CLIENT_ID=xxxxxxxxxxxxxxxxxxxxxxxxxxx
    FGA_CLIENT_SECRET=xxxxxxxxxx-xxxxxxxxxxxxxxxxxxxxxxxxxxx
    # Required only for non-US regions
    FGA_API_URL=https://api.xxx.fga.dev
    FGA_API_AUDIENCE=https://api.xxx.fga.dev/

Obtain OpenAI API Key

Use this page for instructions on how to find your OpenAI API key. Once you have your key, update the .env file accordingly.

Configure Auth0 FGA

1. Create a client

   Navigate to Settings and in the Authorized Clients section click + Create Client button. On the new page give your client a name and mark all three client permissions then click Create.

2. Copy the information on the modal and update your .env file with the values you now have for FGA_STORE_ID, FGA_CLIENT_ID, and FGA_CLIENT_SECRET. Click Continue to get values for FGA_API_URL and FGA_API_AUDIENCE.

How to run it

1. Install dependencies using Poetry.

   $ python -m venv venv
   $ source venv/bin/activate
   $ pip install -r requirements.txt

2. Initialize the FGA model and tuples

   $ python ./scripts/fga_init.py

3. Running the example

   python main.py

---

Auth0 is an easy to implement, adaptable authentication and authorization platform. To learn more checkout Why Auth0?

This project is licensed under the Apache 2.0 license. See the LICENSE file for more info.