<div align="center">

🤖 MAIRS: Multi-Agent Intelligent Research System

An advanced AI-powered research platform leveraging multi-agent systems to conduct comprehensive investigations and generate professional reports

Features • Tech Stack • Installation • Usage • Architecture

---

</div>

📋 Overview

MAIRS (Multi-Agent Intelligent Research System) is a cutting-edge research automation platform that harnesses the power of collaborative AI agents to revolutionize how research is conducted. Built on CrewAI and powered by Groq's lightning-fast inference engine, MAIRS orchestrates specialized AI agents that work together seamlessly to gather, analyze, and synthesize information into comprehensive, publication-ready reports.

Whether you're a researcher exploring emerging technologies, an analyst investigating market trends, or a professional seeking deep insights on complex topics, MAIRS streamlines your research workflow from initial data collection to final report generation — all through an intuitive, interactive interface.

✨ Key Features

🔍 Automated Multi-Source Research

Leverages SerperDevTool to collect current, reliable data from diverse web sources, ensuring comprehensive coverage of your research topic.

📊 Advanced AI-Driven Analysis

Employs sophisticated data processing algorithms to identify patterns, trends, correlations, and key insights from collected information.

📝 Professional Report Generation

Produces structured, publication-ready reports complete with:

• Executive summaries
• Detailed findings
• In-depth analysis sections
• Proper citations and references
• Markdown formatting for easy conversion

🖥️ Interactive User Interface

Beautiful Streamlit-powered interface featuring:

• Real-time progress tracking
• Live status updates
• Tabbed result viewing
• Download functionality for all outputs

🛠️ Intelligent Multi-Agent Workflow

Coordinates three specialized AI agents:

• Info Collector - Research specialist for data gathering
• Data Processor - Analysis expert for insight extraction
• Report Generator - Content writer for professional documentation

⬇️ Flexible Output Management

Download individual components (findings, analysis, final report) in Markdown format for presentations, publications, or further processing.

🔒 Secure API Management

Environment-based configuration system for secure API key storage and management.

⚡ High-Performance Inference

Powered by Groq's cutting-edge inference technology for rapid response times and efficient processing.

🛠️ Tech Stack

<div align="center">

Core Technologies

| Technology | Purpose | Version | |------------|---------|---------| | | Primary programming language | 3.8+ | | | Web framework for interactive UI | Latest | | | Multi-agent orchestration framework | Latest | | | High-performance AI inference engine | Latest |

Tools & Integrations

| Tool | Purpose | |------|---------| | SerperDevTool | Real-time web search and data collection | | FileReadTool | Reading and processing Markdown files | | FileWriterTool | Writing and saving research outputs | | Python-dotenv | Secure environment variable management | | Markdown | Structured document formatting |

</div>

🏗️ System Architecture

┌─────────────────────────────────────────────────────────────┐
│                   STREAMLIT USER INTERFACE                  │
│  ┌──────────────┐  ┌──────────────┐  ┌──────────────────┐ │
│  │ Input Topic  │  │  Progress    │  │  Results Tabs    │ │
│  │   Form       │  │  Tracking    │  │  - Findings      │ │
│  │              │  │              │  │  - Analysis      │ │
│  │              │  │              │  │  - Report        │ │
│  └──────────────┘  └──────────────┘  └──────────────────┘ │
└─────────────────────────────┬───────────────────────────────┘
                              │
                       app.py (Main)
                              │
┌─────────────────────────────▼───────────────────────────────┐
│                  CREWAI ORCHESTRATION LAYER                 │
│                         (crew.py)                           │
│                                                             │
│  ┌──────────────────────────────────────────────────────┐  │
│  │            MULTI-AGENT COORDINATION                  │  │
│  │                                                      │  │
│  │  ┌─────────────┐  ┌─────────────┐  ┌─────────────┐ │  │
│  │  │   Agent 1   │  │   Agent 2   │  │   Agent 3   │ │  │
│  │  │    Info     │→ │    Data     │→ │   Report    │ │  │
│  │  │  Collector  │  │  Processor  │  │  Generator  │ │  │
│  │  └─────────────┘  └─────────────┘  └─────────────┘ │  │
│  │       ↓                 ↓                 ↓         │  │
│  │  ┌─────────────┐  ┌─────────────┐  ┌─────────────┐ │  │
│  │  │  Research   │  │  Analysis   │  │   Writing   │ │  │
│  │  │    Task     │  │    Task     │  │    Task     │ │  │
│  │  └─────────────┘  └─────────────┘  └─────────────┘ │  │
│  └──────────────────────────────────────────────────────┘  │
└──────────────┬────────────────────────────────┬─────────────┘
               │                                │
        ┌──────▼──────┐                  ┌─────▼──────┐
        │   Groq AI   │                  │  Serper    │
        │   Engine    │                  │   Search   │
        │  (LLM-70b)  │                  │    API     │
        └─────────────┘                  └────────────┘
                │                                │
                └────────────┬───────────────────┘
                             │
                   ┌─────────▼──────────┐
                   │  Output Files      │
                   │  - findings.md     │
                   │  - analysis.md     │
                   │  - final_report.md │
                   └────────────────────┘

🎯 How It Works

Phase 1: Data Collection 🔍

The Info Collector agent initiates comprehensive web searches using SerperDevTool, gathering current and reliable information from multiple sources relevant to your research topic.

Phase 2: Intelligent Analysis 📊

The Data Processor agent analyzes collected data, identifying:

• Key patterns and trends
• Statistical correlations
• Important insights
• Notable findings
• Emerging themes

Phase 3: Report Synthesis 📝

The Report Generator agent compiles findings into a professional document with:

• Executive summary
• Methodology section
• Detailed findings
• Comprehensive analysis
• Conclusions and recommendations
• Properly formatted citations

📥 Installation

Prerequisites

• Python 3.8 or higher
• pip package manager
• Git
• API Keys:
  • Serper API Key (Get it here)
  • Groq API Key (Get it here)

Step-by-Step Setup

1. Clone the repository

git clone https://github.com/PasinduSuraweera/MAIRS-Multi-Agent-Intelligent-Research-System.git
cd MAIRS-Multi-Agent-Intelligent-Research-System

1. Create a virtual environment (recommended)

python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

1. Install dependencies

pip install -r requirements.txt

1. Configure environment variables

Create a .env file in the project root:

SERPER_API_KEY=your_serper_api_key_here
GROQ_API_KEY=your_groq_api_key_here

1. Run the application

streamlit run app.py

1. Access the interface Open your browser and navigate to:

http://localhost:8501

📦 Required Dependencies

Create a requirements.txt file with:

streamlit>=1.28.0
crewai>=0.1.0
groq>=0.4.0
python-dotenv>=1.0.0
langchain>=0.1.0
langchain-groq>=0.0.1

🚀 How to Use

Step 1: Launch the Application

Start the Streamlit server and access the web interface through your browser.

Step 2: Configure API Keys 🔑

Ensure your .env file contains valid API keys:

SERPER_API_KEY=your_actual_serper_key
GROQ_API_KEY=your_actual_groq_key

Step 3: Input Research Topic 📝

Enter your research subject in the text input field. Examples:

• "Artificial Intelligence developments in 2025"
• "Climate change impact on agriculture"
• "Blockchain technology in healthcare"
• "Quantum computing breakthroughs"

Step 4: Begin Investigation 🚀

Click the "Begin Investigation" button to start the multi-agent research process.

Step 5: Monitor Progress 📊

Watch real-time updates as each agent completes its tasks:

• ✅ Research data collection
• ✅ Analysis processing
• ✅ Report generation

Step 6: Review Results 📄

Navigate through three comprehensive tabs:

Initial Findings

• Raw research data
• Source information
• Key facts discovered

Detailed Analysis

• Pattern identification
• Trend analysis
• Insight extraction
• Statistical observations

Complete Report

• Executive summary
• Full methodology
• Comprehensive findings
• In-depth analysis
• Conclusions and recommendations
• Citations and references

Step 7: Download Outputs ⬇️

Save your research in Markdown format:

• findings.md - Initial research data
• analysis.md - Detailed analysis
• final_report.md - Complete report

💡 Example Workflow

# Example research topic
topic = "Impact of Large Language Models on Education in 2025"

# MAIRS Process:
# 1. Info Collector searches educational databases, news, research papers
# 2. Data Processor identifies trends in AI adoption, student outcomes, challenges
# 3. Report Generator creates comprehensive 10-page report with citations

# Output: Professional report ready for presentation or publication

📁 Project Structure

MAIRS-Multi-Agent-Intelligent-Research-System/
│
├── app.py                          # Main Streamlit application
├── crew.py                         # CrewAI orchestration configuration
│
├── agents/                         # Agent configurations
│   ├── __init__.py
│   ├── research_specialist.py     # Data collection agent
│   ├── data_analyst.py            # Analysis agent
│   └── content_writer.py          # Report generation agent
│
├── tasks/                          # Task definitions
│   ├── __init__.py
│   ├── research_task.py           # Research task configuration
│   ├── analysis_task.py           # Analysis task configuration
│   └── writing_task.py            # Writing task configuration
│
├── tools/                          # Custom tools and utilities
│   ├── __init__.py
│   └── serper_tool.py             # Web search integration
│
├── outputs/                        # Generated reports
│   ├── findings.md
│   ├── analysis.md
│   └── final_report.md
│
├── .env                            # Environment variables (not in git)
├── .env.example                    # Environment template
├── requirements.txt                # Python dependencies
├── .gitignore                      # Git ignore rules
└── README.md                       # Project documentation

🎨 User Interface Design

MAIRS features a clean, modern interface built with Streamlit:

Navigation & Layout

• Sidebar for configuration and settings
• Main content area for input and results
• Progress indicators for real-time feedback

Visual Elements

• Color-coded status messages
• Progress bars for task completion
• Tabbed interface for organized results
• Download buttons for easy file access

Responsive Design

• Adapts to different screen sizes
• Mobile-friendly interface
• Accessible color schemes

🔒 Security Best Practices

• ✅ API keys stored in .env files (never in code)
• ✅ .env added to .gitignore
• ✅ Environment variable validation
• ✅ Secure API communication
• ✅ No sensitive data in outputs

⚡ Performance Optimizations

• 🚀 Groq Integration - Sub-second inference times
• 🔄 Async Processing - Parallel agent execution where possible
• 💾 Caching - Streamlit caching for repeated queries
• 📦 Efficient File I/O - Optimized read/write operations

🔮 Roadmap & Future Features

🚧 In Development

• [ ] 📊 Data Visualization - Charts and graphs in reports
• [ ] 🌐 Multiple Languages - Support for non-English research
• [ ] 💾 Database Integration - Store research history
• [ ] 📧 Email Reports - Automated delivery of completed research

💡 Planned Features

• [ ] 🤝 Collaboration Tools - Share research with team members
• [ ] 🎨 Custom Templates - Define your own report formats
• [ ] 🔗 API Access - Programmatic access to MAIRS capabilities
• [ ] 📱 Mobile App - iOS and Android applications
• [ ] 🎯 Advanced Filtering - Fine-tune data sources and credibility
• [ ] 🧠 Learning System - Improve based on user feedback
• [ ] 📚 Citation Styles - APA, MLA, Chicago formatting
• [ ] 🔍 Source Verification - Automatic fact-checking integration

🐛 Known Issues & Limitations

• API rate limits may affect research speed on free tiers
• Complex topics may require multiple iterations
• Very recent events (< 24 hours) may have limited coverage
• Large research topics may take 5-10 minutes to complete

🤝 Contributing

Contributions are welcome! Help us make MAIRS even better.

How to Contribute:

1. Fork the Project
2. Create your Feature Branch

   git checkout -b feature/AmazingFeature
   

3. Commit your Changes

   git commit -m 'Add some AmazingFeature'
   

4. Push to the Branch

   git push origin feature/AmazingFeature
   

5. Open a Pull Request

Contribution Ideas:

• 🐛 Bug fixes and error handling improvements
• 📚 Documentation enhancements
• 🎨 UI/UX improvements
• 🔧 New agent capabilities
• 🌐 Language support
• 📊 Data visualization features

📄 License

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

👤 Author

Pasindu Suraweera

• 🐙 GitHub: @PasinduSuraweera
• 💼 LinkedIn: Connect with me
• 📧 Email: Contact me

🙏 Acknowledgments

• 🤖 CrewAI - Revolutionary multi-agent framework
• ⚡ Groq - Lightning-fast AI inference
• 🔍 Serper - Reliable web search API
• 🎨 Streamlit - Beautiful web app framework
• 🐍 Python - The language that powers it all
• 🌟 The open-source AI community

📞 Support

Need help or have questions?

• 🐛 Report a Bug
• 💡 Request a Feature
• 💬 Ask a Question
• 📖 Read the Docs (Coming Soon)

📊 Use Cases

Academic Research

• Literature reviews
• Topic exploration
• Background research
• Citation gathering

Business Intelligence

• Market analysis
• Competitive research
• Trend identification
• Industry reports

Content Creation

• Article research
• Fact-checking
• Topic ideation
• Source compilation

Personal Learning

• Deep dives into new topics
• Skill development research
• Current events analysis
• Educational exploration

---

<div align="center">

⭐ If you find MAIRS useful, please give it a star! ⭐

Made with 🤖 and ❤️ by Pasindu Suraweera

Revolutionizing research through intelligent automation 🚀

</div>