ResearchMind 🔬⚡

Multi-Agent AI Research System

ResearchMind is a powerful multi-agent AI research assistant inspired by the Deep Research capabilities of modern LLM platforms like ChatGPT, Claude, Gemini, and Perplexity.

The system autonomously:

• searches the web,
• gathers reliable sources,
• scrapes detailed content,
• generates structured research reports,
• and critically reviews its own output using AI agents.

Built using LangChain, Mistral AI, Tavily Search, and Streamlit.

---

🚀 Live Demo

🔗 https://multi-agent-system-96hk5izp6m6fxjxqkd34yv.streamlit.app

---

✨ Features

• 🔎 Autonomous AI Search Agent
• 🌐 Real-time Web Search using Tavily
• 📄 Web Scraping & Content Extraction
• ✍️ AI Research Report Writer
• 🧠 AI Critic / Reviewer Agent
• 📚 Structured Research Pipeline
• 🎨 Modern Cyberpunk UI
• 🌙 Dark Theme Interface
• 📥 Markdown Report Download
• ☁️ Cloud Deployment with Streamlit

---

🧠 How It Works

ResearchMind uses a multi-agent architecture similar to advanced AI research systems.

Pipeline Architecture

User Query
    ↓
Search Agent
    ↓
Reader Agent
    ↓
Writer Chain
    ↓
Critic Chain
    ↓
Final Research Report

---

🤖 Agents

1. Search Agent

Searches the internet for:

• recent information
• trusted sources
• relevant URLs
• research material

Powered by:

• Tavily Search API

---

2. Reader Agent

Scrapes and extracts detailed content from selected web pages.

Capabilities:

• content extraction
• summarization
• relevance filtering

---

3. Writer Chain

Generates:

• introduction
• key findings
• conclusions
• structured research reports
• citations & sources

---

4. Critic Chain

Acts like an AI reviewer.

It:

• scores the report
• identifies weaknesses
• suggests improvements
• evaluates research quality

---

🛠️ Tech Stack

AI / LLM

• LangChain
• Mistral AI
• Tavily Search API

Frontend

• Streamlit
• Custom CSS

Backend

• Python
• Multi-Agent Architecture

Deployment

• Streamlit Community Cloud

---

📸 Screenshots

Home Page

• Futuristic cyberpunk interface
• Interactive research pipeline

Research Generation

• Search agent output

• Reader agent scraping

• Final report generation

• Critic feedback analysis

---

⚡ Installation

1. Clone Repository

git clone https://github.com/YOUR_USERNAME/multi-agent-system.git

cd multi-agent-system

---

2. Create Virtual Environment

uv venv

Activate environment:

Windows

.venv\Scripts\activate

---

3. Install Dependencies

uv pip install -r requirements.txt

---

🔑 Environment Variables

Create .env

MISTRAL_API_KEY=your_key
TAVILY_API_KEY=your_key
OPENWEATHER_API_KEY=your_key

---

▶️ Run Locally

streamlit run app.py

---

☁️ Deployment

This project is deployed using Streamlit Cloud.

To deploy:

1. Push project to GitHub
2. Connect repository to Streamlit Cloud
3. Add secrets in Streamlit settings
4. Deploy

---

📂 Project Structure

multi-agent-system/
│
├── app.py
├── agents.py
├── tools.py
├── requirements.txt
├── styles/
├── assets/
├── README.md
└── .streamlit/

---

🎯 Future Improvements

• Vector Database Memory
• Recursive Research Loops
• Multi-page Deep Scraping
• PDF Export
• Live Streaming Responses
• Citation Grounding
• LangGraph Integration
• Async Agent Execution
• Voice Research Assistant
• AI Research History

---

🌟 Inspiration

Inspired by:

• ChatGPT Deep Research
• Claude AI Research
• Gemini Research Tools
• Perplexity AI
• Autonomous AI Agents

---

⭐ If You Like This Project

Star the repository and share it with others 🚀