eBay MCP Server

eBay MCP Server

Enables AI assistants to search eBay listings, track item prices over time, and identify deals below market value using eBay's APIs. It provides tools for category browsing and retrieving detailed information for specific listings.

Category
访问服务器

README

eBay MCP Server

A Model Context Protocol (MCP) server for interacting with eBay's APIs. Enables AI assistants to search eBay listings, track prices, and find deals on homelab equipment and other items.

Features

✅ Implemented (8 tools)

  • 🔍 Search eBay listings - Basic and advanced search with all filters
  • 💰 Price tracking - SQLite database tracks prices over time
  • 📊 Deal detection - Analyzes market data to find bargains (15%+ below average)
  • 📦 Item details - Comprehensive info including seller reputation
  • 📈 Price history - View trends with statistics (min/max/avg/median)
  • 📋 Watchlist - Track multiple items with custom alerts

🚧 Planned (12 tools)

  • 🏷️ Category browsing - Navigate eBay's category hierarchy
  • 🆕 New listings - Fresh deals (last 6-24 hours)
  • Ending soon - Auction sniper helper
  • 📊 Market value - Price research tool

Status

🟢 MVP Complete - 8 core tools implemented and ready for testing!

See IMPLEMENTATION_STATUS.md for detailed progress.

Planned Tools

  • search_ebay - Search listings with filters (keywords, price, condition, category)
  • get_item_details - Get detailed information about a specific listing
  • track_price - Add an item to price tracking
  • get_price_history - View historical price data for tracked items
  • find_deals - Search for items below market value based on historical data

Requirements

  • Python 3.10+
  • eBay Developer Account (for API credentials)
  • MCP-compatible client (Claude Desktop, Cline, etc.)

Quick Start

  1. Get eBay API Credentials

    • Sign up at https://developer.ebay.com/
    • Create an application
    • Copy your App ID
  2. Install

    git clone https://github.com/hanku4u/ebay-mcp-server.git
    cd ebay-mcp-server
    pip install -e .
    
  3. Configure

    cp .env.example .env
    # Edit .env and add your EBAY_APP_ID
    
  4. Test

    python -m ebay_mcp
    # Server will start in stdio mode (MCP standard)
    

Configuration

Set your eBay API credentials as environment variables:

export EBAY_APP_ID="your-app-id"
export EBAY_CERT_ID="your-cert-id"
export EBAY_DEV_ID="your-dev-id"

Or create a .env file:

EBAY_APP_ID=your-app-id
EBAY_CERT_ID=your-cert-id
EBAY_DEV_ID=your-dev-id

Usage

With Claude Desktop

Add to your Claude Desktop config (~/Library/Application Support/Claude/claude_desktop_config.json on macOS):

{
  "mcpServers": {
    "ebay": {
      "command": "python",
      "args": ["-m", "ebay_mcp"],
      "env": {
        "EBAY_APP_ID": "your-app-id",
        "EBAY_CERT_ID": "your-cert-id",
        "EBAY_DEV_ID": "your-dev-id"
      }
    }
  }
}

Alternatively, use the FastMCP CLI:

```json
{
  "mcpServers": {
    "ebay": {
      "command": "fastmcp",
      "args": ["run", "path/to/ebay_mcp/server.py:mcp"],
      "env": {
        "EBAY_APP_ID": "your-app-id",
        "EBAY_CERT_ID": "your-cert-id",
        "EBAY_DEV_ID": "your-dev-id"
      }
    }
  }
}

With OpenClaw

Coming soon - will integrate with OpenClaw's MCP client capabilities.

HTTP Server (Remote Access)

Run as an HTTP server for remote access:

python -m ebay_mcp
# Or with FastMCP CLI:
fastmcp run src/ebay_mcp/server.py:mcp --transport http --port 8000

Then connect from any MCP client:

from fastmcp import Client

async with Client("http://localhost:8000/mcp") as client:
    result = await client.call_tool("search_ebay", {
        "keywords": "Dell PowerEdge R720",
        "max_price": 500
    })
    print(result)

eBay API Access

To use this server, you'll need eBay API credentials:

  1. Sign up for the eBay Developers Program
  2. Create a new application
  3. Generate your App ID, Cert ID, and Dev ID
  4. Choose the appropriate API: Finding API (for search) or Trading API (for account management)

Recurring Searches with OpenClaw

Example cron job for daily homelab equipment searches:

{
  "schedule": { "kind": "cron", "expr": "0 9 * * *", "tz": "America/Chicago" },
  "payload": {
    "kind": "agentTurn",
    "message": "Search eBay for homelab servers under $500 and summarize the best deals"
  },
  "sessionTarget": "isolated"
}

Architecture

Built using FastMCP - the fast, Pythonic way to build MCP servers. FastMCP powers 70% of MCP servers and makes it easy to create production-ready integrations with minimal boilerplate.

Example: Adding a Tool

Adding new tools is as simple as decorating a function:

from fastmcp import FastMCP

mcp = FastMCP("eBay MCP Server")

@mcp.tool
def search_ebay(
    keywords: str,
    max_price: float = None,
    condition: str = "New"
) -> dict:
    """Search eBay listings with filters"""
    # Implementation here
    return {...}

if __name__ == "__main__":
    mcp.run()  # Supports stdio and HTTP transports

FastMCP automatically:

  • Generates JSON schemas from type hints
  • Validates inputs with Pydantic
  • Handles MCP protocol serialization
  • Provides both stdio and HTTP transports

License

MIT

Contributing

Contributions welcome! This is an early-stage project focused on homelab deal hunting.

Author

Created by @hanku4u with AI assistance from RockLobster 🦞

推荐服务器

Baidu Map

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。

官方
精选
JavaScript
Playwright MCP Server

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。

官方
精选
TypeScript
Audiense Insights MCP Server

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。

官方
精选
本地
TypeScript
Magic Component Platform (MCP)

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。

官方
精选
本地
TypeScript
VeyraX

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。

官方
精选
本地
Kagi MCP Server

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。

官方
精选
Python
graphlit-mcp-server

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。

官方
精选
TypeScript
mcp-server-qdrant

mcp-server-qdrant

这个仓库展示了如何为向量搜索引擎 Qdrant 创建一个 MCP (Managed Control Plane) 服务器的示例。

官方
精选
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选