mercari-jp-mcp
Enables searching Mercari Japan products with keyword, price range, condition, and status filters, and supports bulk searches from files with automatic translation to English.
README
Mercari JP Search + Yahoo Auctions GUI
A Python application to search Mercari Japan and Yahoo Auctions products with a Web GUI. Supports simple, bulk, and combined cross-platform searches with powerful filtering options.
Features
- Mercari Simple Search — Single keyword with price range, exclude keywords, status/condition filters
- Mercari Bulk Search — Upload a
.txt,.csv, or.xlsxfile (or paste keywords) and search multiple Mercari keywords at once - Yahoo Auctions Simple Search — Search Yahoo Auctions with price range, status (live/sold), condition, sort, and BIN-only filters
- Yahoo Auctions Bulk Search — Multi-keyword Yahoo Auctions search with CSV export
- Combined Mercari + Yahoo Bulk Search — Search both platforms simultaneously with per-keyword results shown side-by-side
- Filters — Price range, item status, condition, seller review count (Mercari bulk), sort order, BIN-only (Yahoo)
- Shop Item Detection — Automatically detects
/shops/product/vs/item/URLs - Description Viewer — View item descriptions with auto-translate from Japanese to English
- CSV Export — Download search results as CSV for all search types
- Japanese Translation — Product names and descriptions automatically translated to English
Requirements
- Python 3.11+
- Dependencies listed in
pyproject.toml
Setup
-
Clone the repository:
git clone https://github.com/rafayqazi/mercari-jp-mcp.git cd mercari-jp-mcp -
Create and activate a virtual environment:
uv venv source .venv/bin/activate # Windows: .venv\Scripts\activate uv pip install -e .
Usage
Web GUI (Flask Application — Recommended)
Run the built-in Flask web server:
uv run mercari_gui.py
Then open http://127.0.0.1:5000 in your browser.
Windows users can also double-click start_gui.bat to launch directly.
The GUI has five tabs:
- Simple Search — Single keyword with price range, exclude keywords, status/condition filters
- Mercari Bulk Search — Paste multiple keywords (one per line) or upload a
.txt/.csv/.xlsxfile. Filter by status, condition, and seller review count (min/max). Results are grouped by keyword and can be exported as CSV. - Yahoo Auctions — Single keyword search on Yahoo Auctions with price range, status (live/sold), condition, sort order, and BIN-only filter
- Yahoo Bulk — Multi-keyword Yahoo Auctions search with file upload support and CSV export
- Mercari + Yahoo Bulk — Search both platforms at once. Results are displayed side-by-side per keyword with Mercari on the left and Yahoo on the right.
Click Description (Mercari) or Detail (Yahoo) on any result to view item details with auto-translate from Japanese to English. Click View to open the item on the respective platform.
MCP Server (for Claude Desktop)
The MCP server exposes a search_mercari_jp tool that returns formatted markdown results:
uv run server.py
Configure in claude_desktop_config.json:
{
"mcpServers": {
"mercari": {
"command": "uv",
"args": [
"--directory",
"C:\\PATH\\TO\\mercari-jp-mcp",
"run",
"server.py"
]
}
}
}
The server automatically limits API pagination (max_items) for faster responses and returns results as markdown with images and links.
Security
- No API keys, tokens, or credentials are stored in this repository
- All search requests go directly to Mercari Japan's public API
- The
.gitignoreexcludes virtual environments, cache files, and local config - Environment files (
.env,.env.*) are excluded — never commit secrets
How It Works
The app uses the mercari library (MIT license) to interface with Mercari Japan's public API. Search results are parsed, filtered, and optionally translated via Google Translate API.
Acknowledgments
- marvinody/mercari — Mercari API client
- jlowin/fastmcp — MCP framework
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。