该仓库包含模型上下文协议（MCP）的实现，MCP是一个连接人工智能系统与外部工具和数据源的开放标准。

关于IgnitionAI

IgnitionAI 是一家开源组织，致力于通过可访问且协作的工具提升人工智能能力。我们相信社区驱动开发的力量，能够打造强大、道德且富有创新性的人工智能解决方案。

仓库内容

这是一个包含多个MCP服务器实现的单仓库：

• azure-storage-mcp：完整的Azure存储集成，支持表存储、Blob存储、服务总线队列和存储队列
• azure-ai-search：Azure AI Search 集成（开发中）
• LinkedIn-mcp：LinkedIn API 与 OAuth 认证的集成
• LinkedIn-mcp-http：LinkedIn MCP 服务器的 HTTP 变体
• fit-mcp：用于健身和营养计算的MCP实现
• node-mcp：简单Node.js MCP实现
• mcp-template：用于创建新MCP服务器的基础模板

发展

先决条件

• Node.js和PNPM
• 服务特定凭证（参见各个服务器文档）

命令

MCP (Model Context Protocol)

This repository contains implementations of the Model Context Protocol (MCP), an open standard that connects AI systems with external tools and data sources.

About IgnitionAI

IgnitionAI is an open-source organization dedicated to advancing AI capabilities through accessible and collaborative tools. We believe in the power of community-driven development to create robust, ethical, and innovative AI solutions.

Repository Contents

This is a monorepo containing multiple MCP server implementations:

• azure-storage-mcp: Complete Azure Storage integration with Table Storage, Blob Storage, Service Bus Queues, and Storage Queues
• azure-ai-search: Azure AI Search integration (in development)
• linkedin-mcp: LinkedIn API integration with OAuth authentication
• linkedin-mcp-http: HTTP variant of LinkedIn MCP server
• fit-mcp: MCP implementation for fitness and nutrition calculations
• node-mcp: Simple Node.js MCP implementation
• mcp-template: Base template for creating new MCP servers

Development

Prerequisites

• Node.js and pnpm
• Service-specific credentials (see individual server documentation)

Commands

• pnpm build - Build all projects
• pnpm start - Start development server
• pnpm start:prod - Start production server
• pnpm inspect - Run MCP inspector for debugging
• docker compose up -d - Start LinkedIn HTTP server

About MCP

The Model Context Protocol (MCP) is a standard that enables AI models to interact with external tools, APIs, and services, extending their capabilities beyond their initial training. These repositories are available for everyone to use, modify, and contribute to.

Configuration Examples

Azure Storage MCP

"azure-storage": {
  "command": "npx",
  "args": ["azure-storage-mcp"],
  "env": {
    "AZURE_STORAGE_CONNECTION_STRING": "your-connection-string"
  }
}

Azure AI Search MCP

"azure-ai-search": {
  "command": "npx",
  "args": ["azure-ai-search-mcp"],
  "env": {
    "AZURE_SEARCH_ENDPOINT": "your-search-endpoint",
    "AZURE_SEARCH_API_KEY": "your-api-key"
  }
}

LinkedIn MCP

"linkedin": {
  "command": "npx",
  "args": ["-y", "linkedin-mcp"],
  "env": {
    "LINKEDIN_CLIENT_ID": "your-client-id",
    "LINKEDIN_CLIENT_SECRET": "your-client-secret",
    "LINKEDIN_REDIRECT_URI": "your-redirect-uri"
  }
}

Fit MCP

"fit-mcp": {
  "command": "npx",
  "args": ["fitmcp"]
}

How to Get LinkedIn API Keys

To obtain the necessary LinkedIn API credentials (LINKEDIN_CLIENT_ID, LINKEDIN_CLIENT_SECRET, and LINKEDIN_REDIRECT_URI), follow these steps:

1. Create a LinkedIn Developer Account
   Visit LinkedIn Developer Portal and log in with your LinkedIn account.

2. Create a New App
   Click Create app, fill out the required details, and submit.

3. Get Client ID and Client Secret