Azure MCP Inventory Server
Enables CRUD operations on inventory items stored in Azure Cosmos DB via MCP tools, allowing creation, reading, updating, and deletion of items.
README
Azure MCP Inventory CRUD Server and Client
This project contains an MCP (Model Context Protocol) server and async client for managing inventory items in Azure Cosmos DB.
Setup
- Install dependencies:
pip install -r requirements.txt
- Create a
.envfile with your Azure Cosmos DB credentials:
COSMOS_ENDPOINT=https://<your-account>.documents.azure.com:443/
COSMOS_KEY=<your-primary-key>
COSMOS_DATABASE=<your-database-name>
Running the Server
Start the MCP server with HTTP transport:
python server.py
The server will run on http://127.0.0.1:8001 by default.
Using the Client
The client.py module provides an InventoryClient class for consuming the server. It uses async/await and httpx for HTTP communication.
Basic Usage
import asyncio
from client import InventoryClient
async def main():
client = InventoryClient(server_url="http://127.0.0.1:8001")
async with client.connect():
# Create an item
result = await client.create_item(
item_id="ITEM001",
name="Laptop",
category="Electronics",
quantity=10,
price=999.99
)
print(result)
# Read an item
result = await client.read_item("ITEM001", "Electronics")
print(result)
# Update quantity
result = await client.update_quantity("ITEM001", "Electronics", 15)
print(result)
# List items in a category
result = await client.list_category_items("Electronics")
print(result)
# Delete an item
result = await client.delete_item("ITEM001", "Electronics")
print(result)
asyncio.run(main())
Running the Example
To run the included example:
python client.py
Client API
InventoryClient
Constructor
__init__(server_url: str = "http://127.0.0.1:8001")- Initialize the client
Methods
All methods are async and require using async with client.connect(): context manager.
async create_item(item_id, name, category, quantity, price)- Create a new itemasync read_item(item_id, category)- Read an item by ID and categoryasync update_quantity(item_id, category, new_quantity)- Update item quantityasync delete_item(item_id, category)- Delete an itemasync list_category_items(category)- List all items in a category
Server API
The server exposes the following tools via MCP:
create_item- Create a new inventory itemread_item- Read an item from the databaseupdate_quantity- Update the quantity of an existing itemdelete_item- Delete an inventory itemlist_category_items- List all items in a specific category
Architecture
- server.py - FastMCP server that manages inventory CRUD operations with Azure Cosmos DB
- client.py - Async HTTP client for consuming the MCP server
- requirements.txt - Python dependencies
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。