Retail MCP Server
About
A Model Context Protocol (MCP) server implementation for retail product information
Details
- Author
- tsynode
- Downloads
- 188
- Categories
- Other
Jump to
- Get product information by ID
- Search products by query and/or category
- List all available product categories
- Check product availability
Install Node.js 18 or higher, clone the repository, run npm install, then start the server with npm start. MCP clients can connect to http://localhost:3000/mcp. Docker installation is also supported.
Retail MCP Server
A Model Context Protocol (MCP) server implementation for retail product information.
Description
This server provides tools for accessing and searching retail product information using the Model Context Protocol (MCP). It's designed to be used with MCP clients like Claude or other AI assistants that support the protocol.
Features
- Get product information by ID
- Search products by query and/or category
- Get a list of all available product categories
- Check product availability
Tools
| Tool Name | Description |
|-----------|-------------|
| get-product | Get information about a specific product by ID |
| search-products | Search for products by query and/or category |
| get-categories | Get a list of all available product categories |
| check-availability | Check if a product is in stock |
Getting Started
Prerequisites
- Node.js 18 or higher
- Docker (optional)
Installation
1. Clone the repository:
git clone https://github.com/yourusername/retail-mcp-server.git
cd retail-mcp-server
2. Install dependencies:
npm install
3. Start the server:
npm start
The server will be running at http://localhost:3000
Using Docker
1. Build the Docker image:
docker build -t retail-mcp-server .
2. Run the container:
docker run -p 3000:3000 -d retail-mcp-server
Usage
MCP Endpoint
MCP clients can connect to:
http://localhost:3000/mcp
API Endpoints
- GET /: Returns server information
- GET /mcp: SSE endpoint for MCP
- POST /mcp: HTTP endpoint for MCP messages
License
MIT
Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.



