vipervs / researchcrew

How to automate the generation of a scientific article using CrewAI

ResearchNET

# Automated Research Article Generation with CrewAI

This project demonstrates how to automate the generation of a scientific article using CrewAI, a framework for orchestrating multi-agent AI collaboration.

Key Features:

• AI-Powered Research and Writing: Leverages multiple AI agents with distinct roles to conduct research, gather information, and craft a comprehensive article.
• Collaborative AI Workflow: Employs CrewAI to coordinate the agents' efforts, ensuring a cohesive and efficient process.
• Customizable Topic Selection: Allows users to specify the desired research topic, enabling tailored content creation.
• Integrated Research and Writing Tools: Utilizes external tools like Google Custom Search and text scraping capabilities for comprehensive data retrieval.
• Scientific Writing Style: Generates articles in a concise, academic writing style, incorporating citations and a bibliography.

Getting Started:

1. Install required libraries:

   pip install crewai langchain langchain-openai dotenv

2. Create a .env file and add your API Information.
3. Run the project:

   python main.py

4. Enter the desired research topic when prompted.

Workflow:

1. Researcher Agent:
   • Generates relevant keywords for the specified topic.
   • Conducts a literature search using Google Custom Search.
2. Scraper Agent:
   • Scrapes text from the provided links using a custom scraping tool.
3. Writer Agent:
   • Retrieves the scraped text.
   • Composes a detailed scientific article with citations and a bibliography.
   • Formats the article in Markdown.

Output:

The project prints the generated article to the console.