Q

Quarkus MCP Agentic

該項目使用Quarkus框架和Model Context Protocol(MCP)實現了一個基於多MCP服務器的智能代理應用,結合LangChain4j提供AI對話、餐廳推薦、Slack通知等功能,支持開發和生產環境部署。
2分
14

What is this application?

This application demonstrates how to build AI-powered agents using Quarkus (a Java framework) and multiple MCP servers. It connects to various services like Brave Search, Google Maps, Slack, and OpenAI to perform complex tasks through natural language prompts.

How does it work?

The application uses LangChain4j to orchestrate multiple MCP services. You interact with it through natural language prompts, and it intelligently decides which services to use to complete your requests.

Use Cases

Ideal for team coordination tasks like finding restaurants that meet dietary needs, scheduling meetings, sending invitations, and creating calendar events - all through simple chat-like interactions.

Key Features

Multi-service IntegrationSeamlessly combines multiple MCP services (Brave Search, Google Maps, Slack, OpenAI) to complete complex tasks
Natural Language InterfaceUnderstands and executes tasks through conversational prompts in plain English
Contextual MemoryRemembers team preferences and previous interactions to provide personalized responses
Automated WorkflowsPerforms multi-step workflows automatically (search → select → notify → schedule)

Pros and Cons

Advantages
Handles complex multi-service workflows with a single prompt
Remembers context between interactions
Open architecture allows adding new services easily
Built-in development UI for testing and debugging
Limitations
Requires API keys for external services
Needs Node.js and container runtime for full functionality
Current version focuses on specific use cases (team coordination)

Getting Started

Install Prerequisites
Install Node.js/npm and Docker/Podman for container support
Set Up API Keys
Create a .env file with your API keys for Brave Search, Google Maps, Slack, and OpenAI
Run in Development Mode
Start the application with live reload enabled
Access the Interface
Open http://localhost:8080 in your browser and use the chat interface

Example Use Cases

Team Lunch CoordinationFind a restaurant that meets team dietary needs, invite members via Slack, and create calendar events
Information RecallAsk about previous decisions and reasoning

Frequently Asked Questions

Do I need all the API keys to try the application?
Can I change the Slack channel it posts to?
Where are generated files (like ICS) stored?
How do I see what services are being used?

Additional Resources

Quarkus Framework
The Java framework used to build this application
Model Context Protocol
Official MCP documentation
LangChain4j Quarkus Extension
Integration between Quarkus and LangChain4j
GitHub Repository
Source code for MCP servers
安裝
複製以下命令到你的Client進行配置
注意:您的密鑰屬於敏感信息,請勿與任何人分享。
N
Notte Browser
已認證
Notte是一個開源的全棧網絡AI代理框架,提供瀏覽器會話、自動化LLM驅動的代理、網頁觀察與操作、憑證管理等功能,旨在將互聯網轉化為代理友好的環境,並通過自然語言描述網站結構,降低LLM的認知負擔。
639
4.5分
B
Bing Search MCP
一個用於集成微軟Bing搜索API的MCP服務器,支持網頁、新聞和圖片搜索功能,為AI助手提供網絡搜索能力。
Python
212
4分
C
Cloudflare
Changesets是一個用於管理多包或單包倉庫版本和發佈的構建工具。
TypeScript
1.5K
5分
E
Eino
Eino是一個專為Golang設計的LLM應用開發框架,旨在通過簡潔、可擴展、可靠且高效的組件抽象和編排能力,簡化AI應用開發流程。它提供豐富的組件庫、強大的圖形編排功能、完整的流處理支持以及高度可擴展的切面機制,覆蓋從開發到部署的全週期工具鏈。
Go
3.4K
5分
M
Modelcontextprotocol
已認證
該項目是一個集成Sonar API的MCP服務器實現,為Claude提供即時網絡搜索能力。包含系統架構、工具配置、Docker部署及多平臺集成指南。
TypeScript
1.1K
5分
S
Serena
Serena是一個強大的開源編碼代理工具包,能夠將LLM轉化為可直接在代碼庫上工作的全功能代理。它提供類似IDE的語義代碼檢索和編輯工具,支持多種編程語言,並可通過MCP協議或Agno框架與多種LLM集成。
Python
761
5分
O
Open Multi Agent Canvas
Open Multi-Agent Canvas是一個開源的多智能體聊天界面,支持在動態對話中管理多個智能體,用於旅行規劃、研究和通用任務處理。
TypeScript
413
4.5分
Z
Zhipu Web Search MCP
項目提供了兩種服務啟動方式(stdio協議和SSE協議)及對應的客戶端配置方法
Python
54
4.5分
精選MCP服務推薦
M
Markdownify MCP
Markdownify是一個多功能文件轉換服務,支持將PDF、圖片、音頻等多種格式及網頁內容轉換為Markdown格式。
TypeScript
1.7K
5分
B
Baidu Map
已認證
百度地圖MCP Server是國內首個兼容MCP協議的地圖服務,提供地理編碼、路線規劃等10個標準化API接口,支持Python和Typescript快速接入,賦能智能體實現地圖相關功能。
Python
697
4.5分
F
Firecrawl MCP Server
Firecrawl MCP Server是一個集成Firecrawl網頁抓取能力的模型上下文協議服務器,提供豐富的網頁抓取、搜索和內容提取功能。
TypeScript
3.8K
5分
S
Sequential Thinking MCP Server
一個基於MCP協議的結構化思維服務器,通過定義思考階段幫助分解複雜問題並生成總結
Python
249
4.5分
C
Context7
Context7 MCP是一個為AI編程助手提供即時、版本特定文檔和代碼示例的服務,通過Model Context Protocol直接集成到提示中,解決LLM使用過時信息的問題。
TypeScript
5.2K
4.7分
E
Edgeone Pages MCP Server
EdgeOne Pages MCP是一個通過MCP協議快速部署HTML內容到EdgeOne Pages並獲取公開URL的服務
TypeScript
246
4.8分
M
Magic MCP
Magic Component Platform (MCP) 是一個AI驅動的UI組件生成工具,通過自然語言描述幫助開發者快速創建現代化UI組件,支持多種IDE集成。
JavaScript
1.7K
5分
N
Notion Api MCP
已認證
一個基於Python的MCP服務器,通過Notion API提供高級待辦事項管理和內容組織功能,實現AI模型與Notion的無縫集成。
Python
113
4.5分
AIbase
智啟未來,您的人工智慧解決方案智庫
© 2025AIbase