InsightForge

InsightForge is a multi-agent, LLM-based project designed to automate the process of researching and writing articles on specified topics. The system employs a researcher agent to gather information using Google Search, followed by a writer agent that composes a coherent blog or article based on the findings. InsightForge provides an efficient, intelligent solution for generating content from scratch, leveraging the latest in large language models (LLMs) and multi-agent collaboration.

Key Features

• Multi-Agent Architecture: The project utilizes a dual-agent system with specialized roles.
  • Researcher Agent: Conducts research using Serper API for Google search, compiles relevant findings, and formulates a conclusion.
  • Writer Agent: Uses the researcher's findings to write a structured and well-articulated article.
• Dynamic Content Generation: Each article is generated based on real-time data gathered by the researcher agent.
• LLM-Driven Agents: Powered by Google’s Gemini model, allowing for natural language comprehension and coherent article generation.
• Serper API Integration: Enables the researcher agent to gather up-to-date information from Google Search, ensuring relevance and accuracy.

Tools and Frameworks

• Crew AI Agent Framework: Used for building, managing, and orchestrating the multi-agent system.
• Serper API: For retrieving real-time information to guide the research phase.
• Google Gemini Model: Powers the LLM-driven responses for both researcher and writer agents.
• Python: Primary language for project scripts.

Getting Started

Prerequisites

• Python 3.10
• API keys for Serper API
• Access to Google Gemini model for agent responses

Installation

1. Clone the Repository

   git clone https://github.com/MaazLab/InsightForge.git
   cd InsightForge

2. Install Required Packages

   pip install -r requirements.txt

3. Set Up Environment Variables

   • Create a .env file in the root directory with the following details:

     SERPER_API_KEY=your_serper_api_key
     GOOGLE_API_KEY=your_google_gemini_api_key
     

Running the Project

1. Start the Research and Writing Process

   python crew.py --topic "Your chosen topic"

   • The researcher agent will conduct searches on the specified topic, summarize findings, and pass these to the writer agent for article composition.

2. View Output

   • The resulting article will be saved in the output/ directory as a text file named after the topic.

Future Enhancements

• Feedback Mechanism: Implement a feedback loop for iterative improvement of the article quality.
• Enhanced Summarization: Refine the researcher agent to create more granular, topic-specific summaries.
• Advanced Style Options: Enable the writer agent to adapt writing style based on the target audience.

Contributing

We welcome contributions to InsightForge! Please create a pull request with any improvements or submit issues for bug reports or feature requests.

License

This project is licensed under the MIT License.

---

InsightForge: Automating research and content creation through collaborative LLM agents.