广告合作
MCP.so
登录

LLM SSE MCP Demo

@nlinhvu

代码仓库

基本信息

分类

AI 与智能体

许可证

MIT

传输方式

stdio

作者

nlinhvu

关于 LLM SSE MCP Demo

This project demonstrates the integration between LLM clients and MCP (Model Context Protocol) servers using Server-Sent Events (SSE) for real-time communication.

LLM SSE MCP Demo 配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "llm-sse-mcp-demo-2025": {
      "command": "docker",
      "args": [
        "compose",
        "up",
        "-d"
      ]
    }
  }
}

LLM SSE MCP Demo 工具

工具列表暂未提供

当前收录暂时没有可展示的工具列表，不代表服务没有工具。请查阅项目文档了解支持的能力。

工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。

概览

What is LLM SSE MCP Demo?

LLM SSE MCP Demo is a multi-application demo project that integrates LLM clients with MCP (Model Context Protocol) servers using Server-Sent Events (SSE) for real-time communication. It includes an OAuth2 authorization server, an MCP server exposing math and date/time tools, and a web-based LLM client supporting Anthropic Claude, OpenAI GPT, Google Gemini, and Ollama. It targets developers building LLM-powered applications with external tool calling and observability.

How to use LLM SSE MCP Demo?

Start the observability stack (Prometheus, Tempo, Loki, Grafana) with Docker Compose, then run each Spring Boot app via ./gradlew bootRun in order: authorization server (port 9000), MCP server (port 8080), LLM client (port 10101). Set environment variables for cloud LLM API keys as needed. Open a browser to http://localhost:10101 to send queries like math operations or time requests.

Key features of LLM SSE MCP Demo

  • OAuth 2.0 authorization for MCP services
  • Mathematical tools: addition and multiplication
  • Date/time tools: current time and alarm setting
  • SSE-based real-time client-server communication
  • Multi-provider LLM support (Claude, GPT, Gemini, Ollama)
  • Integrated observability with Grafana, Prometheus, Tempo, Loki

Use cases of LLM SSE MCP Demo

  • Building LLM applications that can perform calculations and time queries
  • Demonstrating MCP protocol tool discovery and execution with SSE
  • Exploring OAuth2 security for LLM-triggered tool calls
  • Testing multi-model LLM integration in a single chat interface
  • Learning observability setup for MCP-based Spring Boot services

常见问题

1

What LLM models are supported?

Anthropic Claude (claude-sonnet-4-20250514), OpenAI (o4-mini-2025-04-16), Google Gemini (gemini-2.5-flash-preview-05-20), and Ollama (qwen3:8b locally).

2

What tools does the MCP server provide?

Four tools: `sumNumbers(int, int)` – adds two numbers; `multiplyNumbers(int, int)` – multiplies two numbers; `getCurrentDateTime()` – returns current date/time; `setAlarm(String)` – sets an alarm for an ISO-8601 time.

3

How do I start the demo?

Start the Docker Compose services for observability, then run each Spring Boot application in order: authorization server, MCP server, LLM client, using `./gradlew bootRun`. Access the chat at `http://localhost:10101` and Grafana at `http://localhost:3000`.

4

What are the prerequisites?

Java 17 or higher, Gradle, Docker and Docker Compose, API keys for cloud LLM providers (optional), and Ollama installed locally if using local models.

5

How do I access the web interface?

Open `http://localhost:10101` in a browser after starting the LLM client. For Grafana dashboards (Prometheus, Tempo, Loki), use `http://localhost:3000`.

评论

AI 与智能体 分类下的更多 MCP 服务器

Gemina

Gemina Tech Ltd.

Extract, search and tag any document — invoices, receipts, contracts, forms, or your own templates. 13 tools, one sign-in (OAuth 2.1) or API key, free FileTag tier, data residency in the EU, US, Israel or Asia.

Melaya

Melaya Labs

**Melaya gives an AI assistant hands.** It is a remote MCP server, so there is nothing to install and no API key to configure — one URL works in Claude, ChatGPT, Le Chat, Cursor and any MCP client. ### Your Android phon

Poker Bots All Stars

Agent Disco

Live Texas Hold'em for AI agents: your assistant plays your 3D star, talks trash, anyone can watch.

F

Factanker

FACTANKER serves 1.18B+ evidence-backed facts about US companies, banks, nonprofits and federal spending, sourced from SEC EDGAR, FFIEC call reports, IRS 990 filings and USAspending. Every value carries its evidence: fil

Pocket Network Agentic Marketplace: Pay-per-Call APIs

Pocket Network Foundation

MCP server for the Pocket Network agentic marketplace: search our catalog of data services and utilities and call them, paying per request in USDC on Base with x402. No account or API key. Search and describe are free; p

广告合作