🔬 ResearchFlow AI - Multi Agent System

A Multi-Agent AI Research System that autonomously searches the web, extracts relevant information, generates a structured research report, and critically reviews the final output using specialized AI agents.

---

🌐 Live Demo: https://researchflow-ai-system.onrender.com/

Note: Since the application is hosted on Render's free tier, the first request may take 30–60 seconds while the server wakes up.

🚀 Overview

ResearchFlow AI is an intelligent research assistant built with LangChain, LLMs, and Streamlit. Instead of relying on a single AI model, it uses a multi-agent architecture where each agent performs a specialized task in the research pipeline.

The system can:

• Search the web for recent information
• Scrape detailed content from relevant sources
• Generate a structured research report
• Critically evaluate the report and provide feedback

---

Features

• Multi-Agent Architecture
• Web Search using Tavily API
• Intelligent Web Content Extraction
• AI-powered Research Report Generation
• Automated Report Review & Scoring
• Modern Streamlit Interface
• Download Generated Report

---

Architecture

                User Query
                     │
                     ▼
                Search Agent
                     │
                     ▼
                Reader Agent
                     │
                     ▼
                Writer Chain
                     │
                     ▼
                Critic Chain
                     │
                     ▼
              Final Research Report

---

🛠️ Tech Stack

• Python
• LangChain
• Streamlit
• Tavily Search API
• Groq LLM
• BeautifulSoup
• Requests

---

📂 Project Structure

ResearchFlow-AI-System/
│
├── app.py                 # Streamlit UI
├── agents.py              # AI Agents
├── tools.py               # Search & Scraping Tools
├── pipeline.py            # Research Pipeline
├── pyproject.toml         # Project configuration
├── requirements.txt       # Dependencies
├── uv.lock                
├── README.md
└── .gitignore

---

⚙️ Installation

Clone the repository

git clone https://github.com/Swainakash0799/ResearchFlow-AI-System.git
cd ResearchFlow-AI-System

Install dependencies

Using uv

uv sync

Activate on Windows:

.venv\Scripts\activate

---

🔑 Environment Variables

Create a .env file in the project root.

TAVILY_API_KEY=your_tavily_api_key
GROQ_API_KEY=your_groq_api_key

You can obtain the API keys from:

• Groq: https://console.groq.com/
• Tavily: https://tavily.com/

---

▶️ Run the Application

streamlit run app.py

---

🔄 Workflow

1. User enters a research topic.
2. Search Agent gathers recent information.
3. Reader Agent extracts detailed content from reliable sources.
4. Writer Chain generates a structured research report.
5. Critic Chain evaluates the report and provides feedback.
6. User can download the final report.

---

📸 Demo