
mcp
包含用于开发的示例 MCP 服务器,支持 Python 和 TypeScript。
Repository Info
About This Server
包含用于开发的示例 MCP 服务器,支持 Python 和 TypeScript。
Model Context Protocol (MCP) - This server can be integrated with AI applications to provide additional context and capabilities, enabling enhanced AI interactions and functionality.
Documentation
Example MCP Servers
This repo contains example developer MCP Servers.
Initial Setup
The following sections contain setup instructions for Python and TypeScript examples.
Python Servers
Python Setup and Start Instructions
Note: Python projects have the same setup and start instructions:
Setup (one time):
cd <server directory>
uv venv
source .venv/bin/activate
To start:
make dev
1. Claims MCP Server
Test claims based on a pydantic model. Lookup by id.
Directory: mcp-claims-python
See Python Setup and Start Instructions
2. System Configuration Management Database (CMDB) MCP Server
Test systems based on a pydantic model. Lookup by id.
Directory: mcp-cmdb-python
See Python Setup and Start Instructions
3. Prompts MCP Server
Test prompts based on a pydantic model. Search by keywords using role and descriptions.
Directory: mcp-prompts-python
See Python Setup and Start Instructions
4. Provider MCP Server
Test providers based on a pydantic model. Lookup by id.
Directory: mcp-provider-python
See Python Setup and Start Instructions
5. Pharmacy MCP Server
Search a Chroma vector store loaded from a PDF with pharmacy drug information (see pharmacydrugs.pdf). Does a similarity search.
Directory: mcp-vector-search
The vector store must be created before starting the MCP Server. The current implementation used an OpenAI embedding model and LangSmith for observability. You will need a LangSmith API key and an OpenAI API key.
To create the vector store, follow these steps:
cd mcp-vector-search- Copy
.env.exampleto.env - In
.env, fill in the keys for LANGSMITH_API_KEY and OPENAI_API_KEY:
LANGSMITH_API_KEY=your-api-key-if-tracing
OPENAI_API_KEY=your-api-key
-
Oepn the Jupyter notebook load-store.ipynb and run all the cells.
-
See Python Setup and Start Instructions to start the MCP server.
TypeScript Servers
1. Demo TypeScript MCP Server
This is the sample TypeScript MCP Server that has a tool to add two numbers. It is currently configured to run using the stdio transport.
Directory: mcp-demo-typescript
Since this is configured to run with stdio transport, the typescript must be compiled:
cd mcp-demo-typescriptnpm installnpm run build
Quick Start
Clone the repository
git clone https://github.com/duanewood/mcpInstall dependencies
cd mcp
npm installFollow the documentation
Check the repository's README.md file for specific installation and usage instructions.
Repository Details
Recommended MCP Servers
Discord MCP
Enable AI assistants to seamlessly interact with Discord servers, channels, and messages.
Knit MCP
Connect AI agents to 200+ SaaS applications and automate workflows.
Apify MCP Server
Deploy and interact with Apify actors for web scraping and data extraction.
BrowserStack MCP
BrowserStack MCP Server for automated testing across multiple browsers.
Zapier MCP
A Zapier server that provides automation capabilities for various apps.