module-27-langchain4j-chatbot

Module 27: LangChain4j Chatbot Agent (portability flavor)

The module-25 chatbot, built with LangChain4j — the other top Java AI framework. Same ACP agent, different way to reach the model.

Three flavors, one agent

Module	How it calls the model
25	Raw Anthropic Java SDK (clearest call site)
26	Spring AI ChatClient
27 (this)	LangChain4j ChatModel

The ACP skeleton is identical to the echo agent. The only change from echo is the prompt handler:

String answer = model.chat(req.text());   // model is a LangChain4j ChatModel
context.sendMessage(answer);

Switch providers without touching the agent

ChatModel is LangChain4j's provider-agnostic interface. We build an AnthropicChatModel; swap the dependency + builder to use OpenAI, Gemini, or a local Ollama model — the ACP handler doesn't change:

<!-- this module -->
<dependency>
  <groupId>dev.langchain4j</groupId>
  <artifactId>langchain4j-anthropic</artifactId>
</dependency>
<!-- swap for: langchain4j-open-ai / langchain4j-google-ai-gemini / langchain4j-ollama -->

Prerequisites

• Java 17
• ANTHROPIC_API_KEY exported (used by AnthropicChatModel)

Build & run

./mvnw package -pl module-27-langchain4j-chatbot -q
./mvnw exec:java -pl module-27-langchain4j-chatbot

Note: langchain4j-anthropic is fetched from Maven Central on first build.

Conversation memory

This module keeps it to a single turn for clarity. LangChain4j adds memory via MessageWindowChatMemory (or an AiServices interface with @MemoryId) — see the LangChain4j docs.