For structured, book-length reading or offline reference material:
To build a project inspired by the best GitHub blueprints, you need to properly initialize your build system. Spring AI provides dedicated Starters that tie cleanly into the Spring Boot ecosystem. Maven Dependencies
Here's a look at the repository's structure and key resources:
// Storing the chunks in a vector database for efficient retrieval vectorStore.add(splitDocuments); spring ai in action pdf github
Here are the types of projects and patterns you will find in popular GitHub repositories: 1. The ChatClient and Prompt Templates
One of the most common real-world tasks is loading information from PDF files to build a custom RAG pipeline. Spring AI makes this surprisingly simple with its spring-ai-pdf-document-reader module. To use it, you add a dependency and then, as shown in the earlier code, you can load a PDF, split it into chunks, store those chunks in a vector store, and finally query them with an AI model. The spring-ai-cli-chatbot example on GitHub demonstrates this entire flow end-to-end, from reading the PDF to answering questions about it, making it the ideal template to adapt for your own documents.
This write-up explores the core concepts of Spring AI, mimicking the style of an "In Action" technical guide. We will cover project setup, the "Prompt Template" pattern, RAG (Retrieval-Augmented Generation), and how to structure a production-ready repository. The ChatClient and Prompt Templates One of the
Add the following dependencies to your pom.xml file. Ensure you use the correct Spring AI Bill of Materials (BOM) to manage versions.
org.springframework.ai spring-ai-bom 1.0.0-M6 pom import org.springframework.boot spring-boot-starter-web org.springframework.ai spring-ai-openai-spring-boot-starter Use code with caution. Application Configuration
Did you find this guide helpful? Let us know in the comments what specific AI integration you are building with Spring! Use code with caution.
@RestController @RequestMapping("/api/ai") public class ChatController private final ChatClient chatClient; public ChatController(ChatClient.Builder builder) this.chatClient = builder .defaultSystem("You are a helpful and concise software architecture assistant.") .build(); @GetMapping("/generation") public String generateAnswer(@RequestParam(value = "message") String message) return this.chatClient.prompt() .user(message) .call() .content(); Use code with caution. Achieving Structured Output
To maintain context in a multi-turn conversation, Spring AI provides several memory advisors. The MessageChatMemoryAdvisor is a common one:
Start by adding the Spring AI BOM and the OpenAI starter (assuming you are using OpenAI).