🌲 RAG Forest — Context-Aware Knowledge & Presentation Generator

Visit RAGForest

📌 Project Summary

RAG Forest is a context-aware document intelligence system designed to help students, faculty, and professionals rapidly generate:

• 📊 Context-dependent PowerPoint presentations
• 📝 Structured notes
• 🧩 Architecture diagrams, UML diagrams, and flowcharts
• 📚 Comprehensive answers synthesized from multiple documents

Unlike generic LLM-based generators, RAG Forest grounds every output strictly in user-provided documents, ensuring relevance, accuracy, and contextual integrity.

---

📑 Table of Contents

• Project Summary
• Goal of the Project
• Problem Statement
• Target Users
• What Makes RAG Forest Different
• Current Capabilities
• Current Limitations
• How to Run the Project Locally
• LLM Configuration
• Features Under Active Development
• Future Roadmap
• Platform Vision

🎯 Goal of the Project

The core goal of RAG Forest is to:

Reduce the time and effort required to understand a topic by automatically extracting, synthesizing, and structuring knowledge from uploaded documents—while preserving user-defined context.

This project is built around Retrieval-Augmented Generation (RAG) to avoid hallucinated or generic outputs and instead produce document-faithful results.

---

❓ Problem Statement

Traditional approaches to generating notes or presentations using LLMs suffer from key limitations:

• Outputs are often generic and not context-aware
• Documents must be manually read and summarized
• Diagrams and architecture flows require separate tools
• Information from multiple documents is hard to consolidate
• No grounding to what the user actually provided

How RAG Forest Solves This

• Understands user-uploaded documents
• Allows unlimited queries over uploaded content
• Generates answers only from relevant documents
• Produces structured content suitable for:
  • PPTs
  • Notes
  • Diagrams

---

👥 Target Users

• 🎓 Students — exam preparation, concept understanding, notes
• 👩‍🏫 Faculty — lecture slides, structured explanations
• 👨‍💼 Professionals — architecture design, technical summaries

---

🧠 What Makes RAG Forest Different

Feature	Generic LLM Tools	RAG Forest
Context awareness	❌	✅
Multi-document synthesis	❌	✅
Grounded responses	❌	✅
Diagram-ready understanding	❌	✅
PPT & notes oriented	❌	✅

---

✅ Current Capabilities (Implemented)

📄 Document Upload & Understanding

• Upload PDF documents (at least once)
• Upload multiple documents
• Text extraction from PDFs
• Documents indexed and stored for reuse

🔍 Querying & Retrieval

• Ask unlimited queries after upload
• Queries answered using all relevant documents
• Responses are context-dependent, not generic

🧩 Knowledge Synthesis

• Combines information across documents
• Produces structured, readable answers
• Designed to feed into:
  • PPT generation
  • Notes creation
  • Diagram design workflows

🧱 Backend Architecture

• FastAPI-based backend
• Modular RAG pipeline
• Local execution support

---

🚧 Current Limitations (By Design)

• Only text content from PDFs is processed
• Images, tables, and diagrams are not yet interpreted
• Manual LLM API setup required
• UI supports basic interaction (early-stage)

---

🚀 How to Run the Project Locally

🔧 Prerequisites

• Python 3.10+
• Git
• (Optional but recommended) Virtual environment

---

📥 Step 1: Clone the Repository

git clone https://github.com/Sachin-baba-1/RAG_forest.git
cd RAG_forest

Create & Activate Virtual Environment (Optional but Recommended)

python -m venv env
env\Scripts\activate   # Windows

Install Dependencies

pip install -r requirements.txt

Run the Backend

cd forest
fastapi dev backend/main.py

Access the Application

http://127.0.0.1:8000/ui/

---

🔐 LLM Configuration (Current)

• Currently supports Mistral (manual API setup)
• User must configure the API key locally
• Selected for free and accessible experimentation

---

🛠️ Features Under Active Development

• 🔑 User authentication (sign-in system)
• 📂 Document selection per query
• 💬 Multiple chat sessions
• 📑 Custom number of pages for notes and PPTs
• 🎨 Template selection for presentations
• 📊 Structured PPT generation
• 📄 Exportable document notes

---

🔮 Future Roadmap

🧠 Advanced Document Understanding

• Image interpretation inside PDFs
• Table and diagram comprehension
• Context-aware diagram extraction

📐 Diagram & Architecture Design

• UML diagrams
• Architecture flowcharts
• Concept graphs

---

🌍 Platform Vision

• Hosted online platform
• No manual API setup required
• Support for multiple LLM providers
• Plug-and-play API key management