🥈 This project won 2nd place at the Agents Hackathon organised by
Discover, compare, and install AI research repositories with ease!
- 🤖 AI-Powered Discovery: Scans thousands of research papers to find the most relevant repositories
- 📊 Quality Metrics: Filter by citations, GitHub stars, and recent updates
- ⚡ One-Click Installation: Automatic setup of complex research repositories
- 🖥️ Modern Interface: Futuristic design with immersive visual effects (starfield, smooth animations)
- 🌗 Dark/Light Mode: Theme support for optimal user experience
- 🗣️ Multi-Agent Communication: Achieves accurate search through the web, finding the most relevant papers and codebases
[!NOTE] GitHunter is tested with Python $\geq$ 3.10 🐍.
- Clone the repository: 🌀
git clone https://github.com/henribonamy/hackathon_unaite.git cd unaite_def
uv is the recommended package installer and virtual environment manager for this project, offering faster installation and dependency resolution.
Install dependencies and activate the environment: 📦
uv sync
source .venv/bin/activateIf you prefer using pip, follow these steps:
-
(Optional) Create and activate a virtual environment: 🐍
python -m venv venv source venv/bin/activate -
Install dependencies: 📦
pip install -e .
Create a .env file: 🔑
In the project root, create a file named .env and add your Anthropic API key:
ANTHROPIC_API_KEY=your_anthropic_api_key_hereYou can generate an Anthropic API key from the Anthropic Console
If you installed with UV, run the project with:
# Make sure your virtual environment is activated
source .venv/bin/activate
# Start the server
uv run server.pyIf you installed with pip, run the project with:
# Make sure your virtual environment is activated (if you created one)
source venv/bin/activate # Skip if not using a virtual environment
# Start the server
python server.pyAfter running the command, open the link shown in your terminal (usually http://127.0.0.1:1414/) in your web browser. This will open the GitHunter interface.
Watch our video demonstration to see GitHunter in action:
To see the search functionality in action, try entering the following prompt in the search bar:
general-purpose method for deep learning on surfaces such as 3D triangle meshes and point clouds like diffusion-net
This will list papers related to this topic.
After clicking on "install" for a paper, the codebase of the corresponding paper will be installed automatically. ✨
This tool is developed using the smolagents package by Hugging Face 🤗, and leverages advanced language models including Claude Sonnet 3.7 and Qwen 2.54 to provide intelligent search and automation capabilities.
Made with ❤️ for the AI research community!


