digitalapplied / groq-shell-assistant

Groq Shell Assistant

Overview

Groq Shell Assistant is a conversational AI assistant that can interact with users and call external tools to perform tasks.

Setup

Prerequisites

• Python 3.8+
• pip

Clone the Repository

git clone https://github.com/skapadia3214/groq-shell-assistant.git

Install Dependencies

cd groq-shell-assistant
pip install -r requirements.txt

Set up Environment Variable

Create a .env file in the root of the repository with the following contents:

GROQ_API_KEY=<your-groq-api-key>
TAVILY_API_KEY=<your-tavily-api-key> (optional)

• You can get your Groq API KEY here.
• You can get your Tavily API KEY here.

Run the Assistant

python main.py

Optional Arguments

You can run the main.py with some optional arguments:

• --model: Specify the Groq model to use (default: llama-3.1-70b-versatile)
• --messages: Specify initial messages in JSON format
• --current_dir: Specify the current working directory for the shell tool.

Interacting with the Assistant

To interact with the assistant, simply type a message and press enter. The assistant will respond with a helpful response.

Contributing

Contributions are welcome! To contribute, simply fork the repository, make your changes, and submit a pull request.

License

This project is licensed under the Apache License 2.0. See the LICENSE file for details.