This repository contains an intelligent, stateful Adaptive RAG research agent built with LangGraph, LangChain, and Google Gemini. It answers questions by consulting private documents first, dynamically grading context relevance, and falling back to public web search when necessary. The application includes a modern Web UI with real-time state visualization and 3D interaction.
From the README
🤖 Doc-Aware Research Agent An intelligent, stateful Adaptive RAG (Retrieval-Augmented Generation) research agent built with LangGraph, LangChain, and Google Gemini. The agent answers questions by consulting your private document collection first, dynamically grading context relevance, and seamlessly falling back to public web search (Tavily or DuckDuckGo) whenever document knowledge is missing or insufficient. --- 📸 Web Interface Preview The application comes with a modern, glassmorphism Web UI featuring real-time state visualization of the LangGraph execution pipeline, drag-and-drop document indexing, and 3D mouse-tracking interaction: --- 🌟 Key Features • 🧠Adaptive Query...
Skills
- HTML
- Python