🤖 AI Support Assistant

Transform any website into an intelligent support assistant powered by AI agents

A sophisticated Streamlit web application that leverages CrewAI multi-agent framework to provide accurate, context-aware support responses by intelligently scraping and analyzing website content.

✨ Key Features

🧠 Dual-Agent Intelligence

• Support Agent: Analyzes website content and crafts detailed responses
• QA Reviewer: Ensures accuracy and quality of all answers
• Memory System: Maintains conversation context for better interactions

🌐 Smart Content Processing

• Real-time Website Scraping: Extract content from any public website
• Intelligent Text Processing: Clean and structure scraped content for optimal AI analysis
• Multi-format Support: Handle various website structures and content types

💬 Modern Chat Experience

• Intuitive Interface: Clean, responsive chat UI with message history
• Real-time Progress: Visual feedback showing agent workflow steps
• Conversation Memory: Maintains context across multiple questions
• Error Handling: Graceful error management with helpful feedback

🎨 Professional Design

• Custom Styling: Modern gradient themes and smooth animations
• Responsive Layout: Optimized for desktop and mobile devices
• Status Indicators: Visual connection status for API and website
• Progress Tracking: Expandable step-by-step agent activity viewer

🚀 Quick Start

Prerequisites

• Python 3.8 or higher
• OpenAI API key (Get one here)

Installation

1. Clone the repository

   git clone https://github.com/akaTalhakhan/Customer_Support.git
   cd Customer_Support

2. Create virtual environment

   python -m venv .venv
   
   # On Windows
   .venv\Scripts\activate
   
   # On macOS/Linux
   source .venv/bin/activate

3. Install dependencies

   pip install -r requirements.txt

4. Launch the application

   streamlit run main.py

5. Access the app

   • Open your browser to http://localhost:8501
   • Enter your OpenAI API key in the sidebar
   • Provide a website URL to analyze
   • Start asking questions!

📋 Usage Guide

Step 1: Configuration

• API Key: Enter your OpenAI API key in the sidebar (securely handled, not stored)
• Website URL: Provide the website you want the assistant to learn from

Step 2: Interaction

• Ask Questions: Type support questions in the chat interface
• View Progress: Expand the progress tracker to see agent workflow
• Continue Conversation: The assistant remembers context from previous messages

Step 3: Management

• Clear History: Use the sidebar button to reset conversation
• Change Website: Update the URL to switch knowledge sources

🛠️ Technical Architecture

Core Components

├── main.py                 # Main Streamlit application
├── requirements.txt        # Python dependencies
└── README.md              # Project documentation

Key Dependencies

• streamlit: Web application framework
• crewai: Multi-agent AI framework
• beautifulsoup4: HTML parsing and content extraction
• requests: HTTP client for website scraping
• langchain: LLM integration and tooling
• chromadb: Vector database for embeddings

Agent Workflow

1. Content Extraction: Scrape and clean website content
2. Context Building: Combine website data with conversation history
3. Support Analysis: Primary agent generates response using website knowledge
4. Quality Review: Secondary agent validates and improves the response
5. Response Delivery: Final answer presented to user with progress tracking

🔧 Configuration Options

Environment Variables

OPENAI_API_KEY=your_api_key_here
OPENAI_MODEL_NAME=gpt-3.5-turbo  # Default model

Customization

• Model Selection: Modify OPENAI_MODEL_NAME for different GPT models
• Temperature Settings: Adjust creativity vs accuracy in agent configurations
• UI Styling: Customize CSS in the main application file

📝 License

This project is licensed under the MIT License - see the LICENSE file for details.

🙏 Acknowledgments

• CrewAI - Multi-agent AI framework
• Streamlit - Web application framework
• OpenAI - Language model API

📞 Support

• Issues: GitHub Issues
• Discussions: GitHub Discussions
• Email: Contact Developer

---

Made with ❤️ using Streamlit & CrewAI
_{Transform websites into intelligent support assistants}