DEV Community: Alkademy

Spring AI + MCP Bridge Tutorial — Connect External Tool Servers to ChatClient (2026)

Alkademy — Wed, 01 Jul 2026 05:51:50 +0000

🔗 The missing link — You learned @Tool agents in our Spring AI agents tutorial and MCP basics in M9-A. This Spring AI MCP bridge tutorial connects them: your ChatClient calls external MCP tool servers the same way Cursor does.

Model Context Protocol (MCP) is how IDE agents discover tools. Spring AI is how you ship AI in production APIs. The Spring AI MCP bridge lets one tool server serve both — no duplicate integrations.

This hands-on guide is part of the AI Developer Tutorials hub. Time: ~45 minutes. Prerequisites: Spring AI first REST endpoint (M7-A), OpenAI API key, Node.js (for the sample MCP server).

Why bridge MCP and Spring AI?
Architecture
Step 1 — Dependencies
Step 2 — Configure MCP client
Step 3 — ChatClient + ToolCallbackProvider
Step 4 — REST endpoint
Step 5 — Test
Bonus — Expose Java tools as MCP server
Production checklist

Why bridge MCP and Spring AI? {#why-bridge}

Approach	Where tools run	Best for
@Tool (M13, M14)	Inside Spring Boot JVM	CRUD agents tied to your services
MCP server (M9-A)	Separate process / remote host	Shared tools for Cursor + APIs
Spring AI MCP bridge	Spring Boot orchestrates MCP clients	Production agents using ecosystem tools

💡 Real-world pattern: Your team builds an MCP server for internal docs search. Developers use it in Cursor. With this bridge, the same server powers your customer-facing Spring Boot agent — one tool implementation, two consumers.

Related reading: Add AI to Angular CRUD app (M14) for in-process @Tool on an existing app · Function calling from Angular (M8-B) for structured tool UX.

Architecture {#architecture}



┌─────────────────┐     stdio/SSE      ┌──────────────────────┐
│  MCP tool server │ ◄──────────────► │  Spring Boot app      │
│  (Node, Java, …) │                    │  spring-ai-starter-   │
└─────────────────┘                    │  mcp-client           │
                                       │         │             │
                                       │         ▼             │
                                       │  ChatClient + OpenA

---

**Read the full tutorial with all code on munonye.com →**

---

**Full tutorial:** [Spring AI + MCP Bridge Tutorial — Connect External Tool Servers to ChatClient (2026)](https://www.munonye.com/spring-ai-mcp-bridge-tutorial-2026/)

**Kindson Munonye** — [GitHub](https://github.com/KindsonTheGenius) · [LinkedIn](https://www.linkedin.com/in/kindson) · [About](https://www.munonye.com/about/)

Spring AI Agents Tutorial — Build a Tool-Calling AI Assistant (2026)

Alkademy — Wed, 01 Jul 2026 05:46:35 +0000

🔥 Why this matters in 2026 — Agentic AI is the fastest-growing search topic in Java developer communities. Teams want LLMs that do things — query databases, call REST APIs, send emails — not just chat. This Spring AI agents tutorial shows the production pattern: ChatClient + @Tool on Spring Boot 3, the same stack as our Angular CRUD series.

Agentic AI is the defining backend trend of 2026. If you already ship Spring Boot REST APIs and Angular frontends, the next feature users expect is an AI assistant that acts on your data — look up orders, summarize tickets, recommend products — using tool calling, not prompt hacks.

This hands-on guide is part of the AI Developer Tutorials hub. You will build a tool-calling AI agent with Spring AI, step by step, in about 60 minutes.

Prerequisites: Spring AI first REST endpoint (M7-A), Java 17+, OpenAI API key, basic Spring Boot.

What is agentic AI?
Architecture overview
Step 1 — Define @Tool methods
Step 2 — Configure the agent ChatClient
Step 3 — REST API for the agent
Step 4 — Test multi-step prompts
Step 5 — Angular UI (optional)
Production checklist
FAQ

What is agentic AI? {#what-is-agentic-ai}

A Spring AI agent combines:

Component	Role
LLM	Plans steps and interprets user intent
@Tool methods	Your Java functions the model can invoke
ChatClient	Orchestrates the tool-calling loop

Unlike a basic chat endpoint (M7-A tutorial), an agent may execute multiple tool calls before answering — e.g. "What's the status of order 1042 and when will it ship?" → getOrder(1042) → getShippingEstimate(1042) → natural language summary.

💡 Tip: Start with 2–3 narrow tools. Agents fail when you expose twenty vague functions. Same lesson as microservices boundaries — see our Microservices hub.

Architecture overview {#architecture}

User prompt
    │
    ▼
POST /api/agent
    │
    ▼
ChatClient (Spring AI)
    ├──► LLM plans tool use
    ├──► @Tool getOrderStatus(orderId)
    ├──► @Tool listLowStockProducts()
    └──► Final answer to user

Stack alignment: This agent sits beside your existing FriendsAPI-style CRUD backend from the [complete CRUD guide](https://www.munonye.com/angu

Read the full tutorial with all code on munonye.com →

Full tutorial: Spring AI Agents Tutorial — Build a Tool-Calling AI Assistant (2026)

Kindson Munonye — GitHub · LinkedIn · About

Spring AI Tutorial — Your First REST Endpoint with OpenAI (2026)

Alkademy — Wed, 01 Jul 2026 05:46:01 +0000

This Spring AI tutorial step by step walks you through your first LLM-powered REST endpoint on Spring Boot 3. You will call OpenAI from Java using ChatClient, keep API keys out of source code, and verify with curl — the backend foundation for our AI Developer Tutorials series.

If you have built REST APIs with our Angular CRUD + Spring Boot series, you already know the architecture; we are adding an AI layer on top.

Prerequisites: Java 17+, Maven, OpenAI API key, basic Spring Boot REST knowledge.

Time: ~45 minutes.

Architecture

Client (curl / Angular)
    POST /api/chat  { "message": "Hello" }
         ▼
ChatController → ChatClient → OpenAI API
         ▼
    { "reply": "..." }

Step 1 — Create the project

curl https://start.spring.io/starter.zip \
  -d dependencies=web \
  -d javaVersion=17 \
  -d name=ai-chat-api \
  -o ai-chat-api.zip
unzip ai-chat-api.zip && cd ai-chat-api

Add Spring AI BOM and OpenAI starter in pom.xml:

<dependencyManagement>
  <dependencies>
    <dependency>
      <groupId>org.springframework.ai</groupId>
      <artifactId>spring-ai-bom</artifactId>
      <version>1.0.0</version>
      <type>pom</type>
      <scope>import</scope>
    </dependency>
  </dependencies>
</dependencyManagement>

<dependencies>
  <dependency>
    <groupId>org.springframework.ai</groupId>
    <artifactId>spring-ai-openai-spring-boot-starter</artifactId>
  </dependency>
</dependencies>

Step 2 — Configure OpenAI

# application.properties
server.port=8080
spring.ai.openai.api-key=${OPENAI_API_KEY}
spring.ai.openai.chat.options.model=gpt-4o-mini
spring.ai.openai.chat.options.temperature=0.7

Run with:

export OPENAI_API_KEY=sk-...
./mvnw spring-boot:run

Never commit API keys. Use environment variables or a secrets manager in production (see M11-A secure AI tutorial).

Step 3 — Chat REST endpoint

@RestController
@RequestMapping("/api/chat")
@CrossOrigin(origins = "http://localhost:4200")
public class ChatController {

  private final ChatClient chatClient;

  public ChatController(ChatClient.Builder builder) {
    this.chatClient = builder
        .defaultSystem("You are a helpful Java and Spring Boot assistant.")
        .build();
  }

  @PostMapping
  public ChatResponse chat(@RequestBody ChatRequest request) {
    String reply = chatClient.prompt()
        .user(request.message())
        .call()
        .content();
    return new ChatResponse(reply);
  }
}

public record ChatRequest(String message) {}
public record ChatResponse(String reply) {}

Step 4 — Test with curl

curl -s -X POST http://localhost:8080/api/chat \
  -H "Content-Type: application/json" \
  -d '{"message":"Explain Spring Boot in one sentence"}' | jq

Expected: JSON with a reply field containing the model response.

Troubleshooting

Error	Fix
401 from OpenAI	Check `OPENAI_API_KEY` is set and valid
Bean ChatClient missing	Ensure `spring-ai-openai-spring-boot-starter` is on classpath
CORS blocked from Angular	Add `@CrossOrigin` or configure `WebMvcConfigurer`

Next steps

M7-B: [Build a streaming chat UI in Angular 19](https://www.munonye.com/angular-19-streaming-c

Read the full tutorial with all code on munonye.com →

Full tutorial: Spring AI Tutorial — Your First REST Endpoint with OpenAI (2026)

Kindson Munonye — GitHub · LinkedIn · About

Secure AI Features — API Keys, JWT, and Rate Limiting in Spring Boot (2026)

Alkademy — Wed, 01 Jul 2026 05:40:22 +0000

Spring Boot secure OpenAI API patterns for production AI features. Required reading after M7-A in AI Developer Tutorials.

Rules

API keys only in server env / vault
Angular sends JWT, not OpenAI keys
Rate limit per user/IP
Validate and sanitize all prompts

JWT-secured controller

@RestController
@RequestMapping("/api/chat")
public class ChatController {
  @PostMapping
  @PreAuthorize("isAuthenticated()")
  @RateLimiter(name = "chat")
  public ChatResponse chat(@RequestBody ChatRequest req, Authentication auth) {
    // log user id, not full prompt in prod
    return new ChatResponse(chatClient.prompt().user(req.message()).call().content());
  }
}

Link Spring Boot hub for REST fundamentals.

Full tutorial: Secure AI Features — API Keys, JWT, and Rate Limiting in Spring Boot (2026)

Kindson Munonye — GitHub · LinkedIn · About

RAG with Spring Boot — Embeddings and Vector Search Step by Step (2026)

Alkademy — Wed, 01 Jul 2026 05:39:46 +0000

Learn how to build a RAG Spring Boot tutorial pipeline that answers questions from your own documents. This post extends the AI Developer Tutorials series and connects to M7-A Spring AI REST basics.

RAG architecture

Documents → chunk → embed → VectorStore
User question → embed → top-K similar chunks → prompt → LLM → answer

Dependencies

<dependency>
  <groupId>org.springframework.ai</groupId>
  <artifactId>spring-ai-pgvector-store-spring-boot-starter</artifactId>
</dependency>
<dependency>
  <groupId>org.springframework.ai</groupId>
  <artifactId>spring-ai-openai-spring-boot-starter</artifactId>
</dependency>

Ingestion service

@Service
public class DocumentIngestionService {
  private final VectorStore vectorStore;
  private final Resource docsFolder;

  public DocumentIngestionService(VectorStore vectorStore,
      @Value("classpath:docs/") Resource docsFolder) {
    this.vectorStore = vectorStore;
    this.docsFolder = docsFolder;
  }

  public void ingestAll() throws IOException {
    for (Resource file : docsFolder.getFile().listFiles()) {
      String text = Files.readString(file.getFile().toPath());
      List<Document> chunks = split(text, 800, 100);
      vectorStore.add(chunks);
    }
  }

  private List<Document> split(String text, int size, int overlap) {
    List<Document> out = new ArrayList<>();
    for (int i = 0; i < text.length(); i += size - overlap) {
      out.add(new Document(text.substring(i, Math.min(i + size, text.length()))));
    }
    return out;
  }
}

Question endpoint

@PostMapping("/api/ask")
public AnswerResponse ask(@RequestBody QuestionRequest req) {
  List<Document> similar = vectorStore.similaritySearch(req.question(), 5);
  String context = similar.stream().map(Document::getContent).collect(Collectors.joining("\n---\n"));
  String answer = chatClient.prompt()
      .system("Answer only from the context below. Say 'I don't know' if not found.\n" + context)
      .user(req.question())
      .call()
      .content();
  return new AnswerResponse(answer);
}

Next: function calling from Angular

M8-B — Structured JSON from LLMs in Angular

Full tutorial: RAG with Spring Boot — Embeddings and Vector Search Step by Step (2026)

Kindson Munonye — GitHub · LinkedIn · About

Run Local LLMs with Ollama + Spring Boot REST API (2026)

Alkademy — Wed, 01 Jul 2026 05:33:59 +0000

This Ollama Spring Boot tutorial runs local LLMs with no OpenAI bill — ideal for development and air-gapped environments. Part of AI Developer Tutorials.

Install Ollama

curl -fsSL https://ollama.com/install.sh | sh
ollama pull llama3.2
ollama serve

Spring configuration

Replace OpenAI starter with:

<dependency>
  <groupId>org.springframework.ai</groupId>
  <artifactId>spring-ai-ollama-spring-boot-starter</artifactId>
</dependency>

spring.ai.ollama.base-url=http://localhost:11434
spring.ai.ollama.chat.options.model=llama3.2

Your M7-A ChatController works unchanged.

Next: Angular signals for chat state (M10-B)

Full tutorial: Run Local LLMs with Ollama + Spring Boot REST API (2026)

Kindson Munonye — GitHub · LinkedIn · About

Model Context Protocol (MCP) for Developers — Build a Simple Tool Server (2026)

Alkademy — Wed, 01 Jul 2026 05:33:24 +0000

Model Context Protocol (MCP) is a 2026 must-know for developers building AI tooling. This hands-on guide is in our AI Developer Tutorials hub.

MCP roles

Role	Example
Host	Cursor IDE, Claude Desktop
Client	MCP client inside the host
Server	Your Java/Node service exposing tools

Minimal tool server (concept)

// Pseudocode — MCP SDK pattern
server.tool("get_weather", { city: "string" }, async ({ city }) => ({
  content: [{ type: "text", text: `Weather in ${city}: 22°C` }],
}));

Bridge to Spring AI

Your existing @Tool methods in M8-B solve the same problem inside the JVM. MCP generalizes this for IDE agents and polyglot stacks.

Next: Event-driven AI agents with Axon (M9-B)

Full tutorial: Model Context Protocol (MCP) for Developers — Build a Simple Tool Server (2026)

Kindson Munonye — GitHub · LinkedIn · About

Full-Stack AI Starter Kit — Hub Summary (2026)

Alkademy — Wed, 01 Jul 2026 05:27:36 +0000

This full stack AI app tutorial summary links every post in the Modern Dev + AI series on munonye.com.

Learning path

Phase	Posts	Outcome
Foundation	M7-A, M7-B	Chat MVP
RAG	M8-A, M8-B, M11-B	Grounded answers
Advanced	M9-A, M9-B, M10-A	MCP, agents, local LLMs
Production	M11-A, M12-A	Security + tests
Reference	M10-B, M12-B	Signals + this summary

All tutorials

Overview: Spring AI + Angular chat feature

GitHub: munonye-ai-chat-spring-angular

Related hubs: Angular · Spring Boot · Microservices

Full tutorial: Full-Stack AI Starter Kit — Hub Summary (2026)

Kindson Munonye — GitHub · LinkedIn · About

Event-Driven AI Agents with Axon and Spring AI (2026)

Alkademy — Wed, 01 Jul 2026 05:26:59 +0000

Connect event-driven AI microservices using Axon Framework and Spring AI. Start with Axon CQRS tutorial and CQRS intro from the Microservices hub.

Flow

HTTP → CommandGateway → Aggregate → Event
EventHandler → ChatClient (classify / plan) → Follow-up Command

Event handler example

@Component
public class SupportTicketAiHandler {
  private final ChatClient chatClient;
  private final CommandGateway gateway;

  @EventHandler
  public void on(TicketCreatedEvent event) {
    String priority = chatClient.prompt()
        .user("Classify priority 1-5 for: " + event.description())
        .call()
        .content();
    gateway.send(new SetTicketPriorityCommand(event.ticketId(), priority));
  }
}

See AI Developer Tutorials for the full series.

Full tutorial: Event-Driven AI Agents with Axon and Spring AI (2026)

Kindson Munonye — GitHub · LinkedIn · About

Build a Documentation RAG Bot for Your Codebase (2026)

Alkademy — Wed, 01 Jul 2026 00:23:11 +0000

Build a documentation RAG bot for your team — capstone of AI Developer Tutorials. Combines M8-A RAG and M7-B Angular chat.

Project structure

docs-bot-api/     Spring Boot — ingest, vector store, /api/ask
docs-bot-ui/      Angular 19 — chat + citation panel
/data/repo/       Cloned project to index

Citation in responses

Return chunk metadata with answers:

public record AnswerResponse(String answer, List<Citation> sources) {}
public record Citation(String file, int chunkIndex) {}

Display citations in Angular with links to GitHub line numbers.

GitHub: munonye-ai-chat-spring-angular

Full tutorial: Build a Documentation RAG Bot for Your Codebase (2026)

Kindson Munonye — GitHub · LinkedIn · About

Angular Signals for AI Chat State Management (2026)

Alkademy — Wed, 01 Jul 2026 00:22:35 +0000

Angular signals tutorial 2026 for AI chat apps — extend M7-B chat UI with computed and effect. From the Angular tutorials hub.

messages = signal<ChatMessage[]>([]);
lastMessage = computed(() => {
  const m = this.messages();
  return m.length ? m[m.length - 1] : null;
});
isWaiting = computed(() => this.loading());

constructor() {
  effect(() => {
    this.messages(); // trigger scroll when messages change
    queueMicrotask(() => this.scrollToBottom());
  });
}

Cross-link reactive forms guide for validated prompt inputs.

Full tutorial: Angular Signals for AI Chat State Management (2026)

Kindson Munonye — GitHub · LinkedIn · About

Function Calling from Angular — Structured JSON from LLMs (2026)

Alkademy — Wed, 01 Jul 2026 00:16:49 +0000

This tutorial shows Angular function calling against a Spring AI backend with @Tool methods — part of AI Developer Tutorials.

Spring Boot tools

@Component
public class ProductTools {
  @Tool(description = "Look up product price by SKU")
  public ProductPrice getPrice(String sku) {
    return catalog.find(sku).map(p -> new ProductPrice(p.sku(), p.name(), p.price()))
        .orElse(new ProductPrice(sku, "Unknown", 0));
  }
}

public record ProductPrice(String sku, String name, double price) {}

this.chatClient = builder
    .defaultTools(productTools)
    .build();

Angular display

interface ProductPrice {
  sku: string;
  name: string;
  price: number;
}

// In component after chat response:
parseProduct(json: string): ProductPrice | null {
  try { return JSON.parse(json); } catch { return null; }
}

Use @if blocks in the template to render a price card when the assistant returns JSON.

Related: Reactive forms validation for input validation before sending prompts.

Full tutorial: Function Calling from Angular — Structured JSON from LLMs (2026)

Kindson Munonye — GitHub · LinkedIn · About

DEV Community: Alkademy

Spring AI + MCP Bridge Tutorial — Connect External Tool Servers to ChatClient (2026)

Table of contents

Why bridge MCP and Spring AI? {#why-bridge}

Architecture {#architecture}

Spring AI Agents Tutorial — Build a Tool-Calling AI Assistant (2026)

Table of contents

What is agentic AI? {#what-is-agentic-ai}

Architecture overview {#architecture}

Spring AI Tutorial — Your First REST Endpoint with OpenAI (2026)

Architecture

Step 1 — Create the project

Step 2 — Configure OpenAI

Step 3 — Chat REST endpoint

Step 4 — Test with curl

Troubleshooting

Next steps

Secure AI Features — API Keys, JWT, and Rate Limiting in Spring Boot (2026)

Rules

JWT-secured controller

RAG with Spring Boot — Embeddings and Vector Search Step by Step (2026)

RAG architecture

Dependencies

Ingestion service

Question endpoint

Next: function calling from Angular

Run Local LLMs with Ollama + Spring Boot REST API (2026)

Install Ollama

Spring configuration

Model Context Protocol (MCP) for Developers — Build a Simple Tool Server (2026)

MCP roles

Minimal tool server (concept)

Bridge to Spring AI

Full-Stack AI Starter Kit — Hub Summary (2026)

Learning path

All tutorials

Event-Driven AI Agents with Axon and Spring AI (2026)

Flow

Event handler example

Build a Documentation RAG Bot for Your Codebase (2026)

Project structure

Citation in responses

Angular Signals for AI Chat State Management (2026)

Function Calling from Angular — Structured JSON from LLMs (2026)

Spring Boot tools

Angular display