CodeCompass

Overview

CodeCompass is a Chabot and Reccomendation engine that aims to enhance the GitHub experience for developers and learners alike by providing a personalized interface to discover and interact with repositories. Our goal is to make learning and discovery in GitHub intuitive, informative, and tailored to each user's interests and needs.

Personalized Recommendations

Utilizing GitHub Data and Machine Learning algorithms, the system suggests GitHub projects that resonate with the user's historical interactions and preferences.

For a full Recommender demonstration amd model explanation, watch our video.

Interactive Exploration and Learning

The embedded chatbot enhances repository exploration, providing valuable insights and fostering a conducive exploration environment.

For a full Chabot demonstration, watch our video.

Usage

Follow these steps to get CodeCompass up and running on your local machine:

1. Clone the repo:

git clone https://github.com/RecandChat/CodeCompass

2. Change to the project directory:

cd CodeCompass

3. Create and activate a Virtual Environment

4. Install dependencies:

pip install -r requirements.txt

5. Create Secrets Diretory at the root (secrets/)

• github_token: Your GitHub API Token.
• instructions: Your chatbot system prompt.
• openAI_key: Your OpenAI API key.

4. Run:

Chatbot

streamlit run frontend/chatbot/app.py

Recommender:

streamlit run frontend/recommender/app.py

Further Documentation

For more in-depth information, refer to the documentation/Final_Report directory within the repository.

Contributors

• Luca Cuneo
• Gabriel de Olaguibel
• Keti Sulamanidze
• Felix Gomez-Guillamon
• Maud Helen Hovland
• Miranda Drummond