🔍 FactBot AI: Your Intelligent Multi-Agent Fact-Checking System

A glimpse of FactBot AI's intuitive dark-themed interface.

🚀 Overview

Welcome to FactBot AI, a sophisticated multi-agent system designed to deliver comprehensive and rapid fact-checking capabilities. Built with the powerful CrewAI framework and presented through an intuitive Streamlit interface, FactBot AI takes a news article or topic as input, processes it through a network of specialized agents, and provides a concise final verdict, supported by verified articles, detailed reasoning, and actionable recommendations.

This intelligent application is engineered to bring clarity and accuracy to the information landscape, ensuring you receive well-substantiated insights along with transparency into the verification process.

✨ Key Features

• Intelligent Multi-Agent System: Leveraging CrewAI, FactBot AI orchestrates a team of specialized AI agents, each focusing on different aspects of fact-checking to ensure thorough and accurate analysis.

• Comprehensive Output: For every input, FactBot AI provides:

  • Final Verdict: A clear and concise judgment on the veracity of the input news/topic.
  • Supporting Articles: Links and summaries of verified sources that underpin the verdict.
  • Detailed Reasoning: A transparent explanation of how the verdict was reached, highlighting key findings from the agents.

• Transparency & Metrics:

  • Sources Verified: Clearly displays the number of external sources consulted and verified by the agents during the fact-checking process.
  • Execution Time: Provides the time taken for the entire fact-checking process, demonstrating the efficiency of the multi-agent architecture.

• User-Friendly Interface: A sleek, dark-themed Streamlit application designed for ease of use and optimal readability.

• Rapid Processing: Experience swift fact-checking, delivering comprehensive results in a minimal timeframe.

🛠️ Technical Stack & Dependencies

FactBot AI is built upon robust and modern technologies:

• Python 3.8+ 🐍: The core programming language.
• CrewAI: A cutting-edge Agentic AI Framework that powers the multi-agent architecture.
• Streamlit 🌐: Provides the interactive and user-friendly web interface.

To run this project, ensure you have these installed.

🔑 API Credentials

FactBot AI relies on external APIs for its comprehensive fact-checking capabilities:

• LLM Key (e.g., Gemini API Key): Essential for the language model's reasoning and understanding.
  • Obtain your Gemini API Key by following the instructions here.
• Serper API Key: Used for efficient and accurate web searching to gather supporting articles.
  • Create a free account at Serper.dev to obtain your API key.

Ensure these API keys are configured as environment variables or within a .env file for secure access.

🚀 Getting Started

Follow these simple steps to set up and run FactBot AI on your local machine:

1. Clone the Repository:

   git clone https://github.com/saikiran1224/FactBot-AI.git
   cd factbot-ai

2. Install Dependencies:

   pip install -r requirements.txt

3. Configure API Keys: Create a .env file in the root directory of the project and add your API keys:

   GEMINI_API_KEY="your_gemini_api_key_here"
   SERPER_API_KEY="your_serper_api_key_here"
   

   Replace "your_gemini_api_key_here" and "your_serper_api_key_here" with your actual API keys.

4. Run the Application:

   streamlit run main.py

5. Access the App: If not automatically opened, Open your web browser and navigate to http://localhost:8501

🎨 Interface & Theming

FactBot AI features a carefully designed dark theme for an optimal viewing experience. The theme is configured in main.py and can be customized:

• Primary Color: #4CAF50 (A trust-evoking green)
• Backgrounds: #121212 (Main), #1E1E1E (Component)
• Text Colors: #E0E0E0 (Light), #A0A0A0 (Secondary)

Feel free to modify these values to suit your aesthetic preferences.

📂 Project Structure

The project is organized into logical components for clarity and maintainability:

• main.py: The entry point of the Streamlit application, handling the user interface and interactions.
• trigger_crew.py: Contains the core logic for orchestrating the multi-agent fact-checking process using CrewAI.
• requirements.txt: List of all Python dependencies required for the project.
• README.md: This documentation file.
• .env: (Recommended) File for securely storing API keys and other environment variables.

🤝 Contributing

We welcome contributions to enhance FactBot AI! If you have suggestions, bug reports, or want to contribute code, please feel free to:

• Submit issues on the GitHub repository.
• Open pull requests with your proposed changes.

📜 License

This project is licensed under the MIT License. See the LICENSE file in the repository for full details.