Infinity Agents

An AI agent for paper searching and summarization, now with real-time streaming responses via WebSockets. Built using agno.

Table of Contents

• Introduction
• Features
• Installation
• Usage
• License

Introduction

Infinity Agents is an AI-based tool designed to search for academic papers and provide summaries, built using the agno framework. This repository contains the necessary components to run the Paper AI agent.

This AI Agent is designed for those who need to search and summarize academic papers efficiently.

Features

• Paper AI: Give it a topic, it will search articles using ArXiv, PubMed, and DuckDuckGo, then summarize the most relevant ones for you.
• Code AI: Provide instructions for coding or data analysis tasks (especially bioinformatics related). It breaks down the task, generates Python/Shell code, and executes it. (Note: Currently backend only, UI integration pending).
• Chater: A general conversational AI.

Run it locally

1. Install the required dependencies (including eventlet for WebSockets):

   pip install -r requirements.txt

2. Set the environment variable DEEPSEEK_API_KEY to your DeepSeek API key, or modify the API_KEY variable in app/config.py.

   # In app/config.py
   API_KEY = os.environ.get("DEEPSEEK_API_KEY", "your API key here")

3. Run the Flask-SocketIO application:

   python app/app.py

   The application now uses eventlet for handling WebSocket connections. Access the application at http://127.0.0.1:8080 (or the configured host/port).

Run it in Docker

1. Clone the repository:

   git clone https://github.com/Vist233/Infinity_Bio-Agents.git
   cd Infinity_Bio-Agents

2. Build and run the Docker container:

   # Make sure to set your API key, e.g., by editing app/config.py before building
   # Or pass it as an environment variable during run time if the app is configured to read it
   docker build -t infinite-agents .
   # The CMD in Dockerfile now runs `python app/app.py` which starts the SocketIO server
   docker run -p 8080:8080 -e DEEPSEEK_API_KEY="your_api_key_here" infinite-agents

   Access the application at http://0.0.0.0:8080.

License

This project is licensed under the Apache-2.0 license.