KSA PDPL GPT Assistant

This repository hosts a specialized chatbot designed to answer questions about the Kingdom of Saudi Arabia's (KSA) Personal Data Protection Law (PDPL). Built with the OpenAI Assistants API and a Streamlit frontend, it serves as an interactive tool to help users understand and navigate PDPL compliance requirements.

Snapshot

Features

• Specialized Knowledge: The assistant is configured to act as a Data Governance and Privacy Consultant specializing in KSA PDPL.
• Context-Aware Responses: If a user refers to an article without specifying a law, the assistant defaults to the KSA "Personal Data Protection Law".
• Interactive Guidance: Suggests relevant follow-up questions to help users explore technical topics and related articles.
• Chat History Management: Allows users to clear the current conversation or export the entire chat to a .txt file for documentation.

How It Works

The application architecture consists of a user-facing frontend and a powerful backend:

• Frontend: A web interface built with Streamlit (chatbot_assistant.py) captures user input and displays the conversation.
• Backend: An OpenAI Assistant is configured with a specific persona, instructions, and knowledge base. The instructions in instructions_for_assistant.txt guide the assistant's behavior, ensuring responses are accurate, focused on data privacy, and helpful.

When a user sends a message, Streamlit forwards it to the OpenAI Assistant, which processes the query based on its instructions and returns a tailored response.

Getting Started

Follow these instructions to set up and run the application locally or with Docker.

Prerequisites

• Python 3.12+
• An OpenAI API Key.
• Docker (for containerized deployment)

Configuration

Before running the application, you need to configure it with your own OpenAI credentials.

1. Create an OpenAI Assistant:

   • Go to the OpenAI Assistants page.
   • Create a new Assistant.
   • For the Instructions, copy and paste the content from instructions_for_assistant.txt.
   • Choose a model (e.g., gpt-4o-mini).
   • Save the Assistant and copy its Assistant ID (e.g., asst_...).

2. Update the Source Code:

   • Open chatbot_assistant.py in a text editor.
   • Replace the placeholder ASSISTANT_ID with your new Assistant ID.
   • Replace the placeholder api_key with your OpenAI API key.

   # chatbot_assistant.py
   
   # Replace with your Assistant ID
   ASSISTANT_ID = 'YOUR_ASSISTANT_ID_HERE' 
   # Replace with your OpenAI API Key
   client = OpenAI(api_key="YOUR_OPENAI_API_KEY_HERE") 

Running Locally

1. Clone the repository:

   git clone https://github.com/mohsinmshabbir/pdpl-gpt-assistant.git
   cd pdpl-gpt-assistant

2. Install dependencies:

   pip install -r requirements.txt

3. Run the Streamlit application:

   streamlit run chatbot_assistant.py

   The application will be available at http://localhost:8501.

Running with Docker

1. Clone the repository:

   git clone https://github.com/mohsinmshabbir/pdpl-gpt-assistant.git
   cd pdpl-gpt-assistant

2. Build the Docker image:

   docker build -t pdpl-gpt-assistant .

3. Run the Docker container:

   docker run -p 8501:8501 pdpl-gpt-assistant

   Access the application in your browser at http://localhost:8501.

File Descriptions

• chatbot_assistant.py: The main Streamlit application file that creates the UI and handles communication with the OpenAI API.
• new_chatbot.py: An alternative version of the chatbot with experimental message editing functionality.
• instructions_for_assistant.txt: A text file containing the core instructions, persona, and response guidelines for the OpenAI Assistant.
• requirements.txt: A list of Python packages required to run the application.
• dockerfile: Instructions for building a Docker container to run the application in an isolated environment.