README.md

🤖 AI Agent Project

✨ Project Overview

AI Agent is an AI-powered question-answering system built with Spring Boot (backend) and Angular 20 (frontend).
It can:

• 💬 Answer user queries.
• 📄 Perform semantic searches on PDF documents.
• 🛠️ Provide custom tools (e.g., retrieve employee information) via a vector store.

---

Features

Backend

• Spring Boot REST API
• AI Agent Integration using Spring AI ChatClient
• Document Indexing & Semantic Search:
  • 📄 PDF Reader: PagePdfDocumentReader
  • ✂️ Text Splitter: TokenTextSplitter
  • 🗄️ Embeddings Storage: SimpleVectorStore
• Custom Tools:
  • 👤 Get employee information
  • 🔍 Search document context
• Endpoints:
  • GET /askAgent?query=<query> — Send a query to the AI agent
  • POST /loadFile — Upload a PDF for semantic search

Frontend

• 🌐 Angular 20 standalone components
• 💬 Simple chat interface
• ⏳ Streaming responses with progress indicator
• 📝 Markdown support for formatted responses
• ⚡ Communicates with backend using HttpClient

---

🏗️ Architecture

Backend Structure

Front Structure

---

Prerequisites

• Java 17+ / JDK compatible with Spring Boot
• Maven
• Node.js 20+ / npm
• Angular CLI 20+
• OpenAI API Key

Usage

• Open the frontend in a browser
• Type your query and click Send
• AI Agent responds with streaming feedback and spinner
• Upload PDFs via /loadFile endpoint to enhance answers

---

Notes

• Maximum file upload: 30MB
• Document embeddings stored in store.json
• Future improvements:
  • Advanced AI tools
  • Multi-component UI
  • Richer UI/UX design

---

License

Open-source for educational and development purposes