🧠 IntelGenie - Multi-Agent Research Assistant

IntelGenie is a LangGraph + LLaMA-powered research assistant that automates research, summarization, and strategic recommendations into a downloadable .docx report — all running locally.

🚀 Features

• Multi-agent pipeline: Research → Summarize → Recommend
• Uses Ollama's LLaMA 3 — no API key needed
• Outputs clean Word reports with cover page
• Smart LLM-generated report filenames

🛠️ Technologies

• LangGraph
• LangChain
• Streamlit
• Ollama
• Python 3.10+

📁 Final Project Directory: IntelGenie/ bash Copy code IntelGenie/ ├── app.py # Streamlit frontend + report download ├── .env.example # Environment variable template (NO SECRETS) ├── requirements.txt # Python dependencies ├── README.md # Project overview and usage instructions │ ├── workflows/ # LangGraph pipeline logic │ └── graph.py │ ├── agents/ # Individual agents for each task │ ├── researcher.py │ ├── summarizer.py │ └── recommender.py │ ├── utils/ # Supporting utilities (DOCX formatting, etc.) │ └── formatting.py │ ├── assets/ # Optional: demo screenshots, sample reports │ ├── demo_screenshot.png │ └── sample_report.docx │ └── LICENSE # (optional) MIT or Apache license

📦 Installation

git clone https://github.com/Chetansonawane7/-IntelGenie---Multi-Agent-Research-Assistant.git
cd IntelGenie
pip install -r requirements.txt

Start Ollama if not running

Ollama run llama3

Start the App:

streamlit run app.py

📁 Example Output Query Filename "How does AI impact healthcare?" AI_in_Healthcare_IntelGenie_Report.docx

📄 License MIT

👤 Built by Your Name

yaml Copy Edit

---

📦 2. requirements.txt

Use this as a base:

streamlit langchain langgraph langchain-community langchain-core python-dotenv duckduckgo-search pypdf faiss-cpu sentence-transformers python-docx

go Copy Edit

Add:

pip freeze > requirements.txt

Then remove your local-only or version-specific junk lines.

🔒 3. .env.example env Copy Edit

If you ever use OpenAI or HuggingFace

OPENAI_API_KEY=your_key_here