CellarTracker MCP

CellarTracker MCP

Connects Claude to your CellarTracker wine cellar, enabling natural language queries for inventory, drinking recommendations, purchase history, and tasting notes.

Category
访问服务器

README

CellarTracker MCP

Connect Claude to your CellarTracker wine cellar. Query your inventory, get drinking recommendations, and analyze purchases and drinking history, all through natural conversation. No need to refresh your inventory, or share your tasting notes. Claude pulls them directly from your CellarTracker account.

Install

Two install methods for different Claude interfaces. You may need one or both:

Desktop Extension (.mcpb) Claude Code Plugin
Works in Chat (Desktop), Cowork Cowork, Code
Tools 8 cellar tools All 10 tools (includes setup-credentials and clear-user-data)
Skills No Yes
Setup One-click download Marketplace or CLI
Credentials Prompted on install, stored in OS keychain Run setup-credentials after install

Which do I need?

  • Chat (Desktop) only → Desktop Extension
  • Code only → Claude Code Plugin
  • Cowork only → Either works (plugin adds skills)
  • Full coverage → Install both

Installing both is safe. The extension covers Chat (Desktop); the plugin covers Cowork and Code. In Cowork, both are accessible without conflict. Note, there is no coverage for claude.ai web chat sessions.

Desktop Extension (Chat & Cowork)

One-click install. No terminal needed.

  1. Download cellartracker-mcp.mcpb from the latest release
  2. Double-click the file to install in Claude Desktop
  3. Enter your CellarTracker username and password when prompted
  4. Start chatting (e.g. "What wines should I open this month?")

Credentials are stored in your OS keychain (macOS Keychain / Windows Credential Manager). To update them later, go to Customize > Connectors > CellarTracker.

Claude Code Plugin (Cowork & Code)

Full experience with tools and skills.

Via Desktop app

  1. Open the Code tab > Customize > Browse plugins > Personal > + > Add marketplace from GitHub > enter slavins-co/cellartracker-mcp
  2. Find "CellarTracker MCP" in the plugin browser and click Install
  3. Set up credentials immediately: In a new Code or Cowork session, say "Set up my CellarTracker credentials"

Via terminal

Step 1: Add the marketplace

/plugin marketplace add slavins-co/cellartracker-mcp

Step 2: Install the plugin

/plugin install cellartracker-mcp@cellartracker-mcp

Step 3: Set up credentials immediately — say:

"Set up my CellarTracker credentials"

Claude will verify and save them. No restart needed. Without credentials, tools will return errors and skills won't have data to work with.

Alternative: Set CT_USERNAME and CT_PASSWORD as environment variables in your shell profile.

Credentials are stored only on your machine. When using the setup tool, they pass through Anthropic's servers as part of the conversation. They are sent to CellarTracker's servers for authentication.

What you can do

Tool What it does
setup-credentials Connect your CellarTracker account (Claude Code plugin only — Desktop Extension handles credentials during install)
search-cellar Find wines by name, color, region, varietal, location, or vintage
drinking-recommendations Wines to open now, sorted by drinking window urgency
cellar-stats Collection overview — totals and breakdowns by any dimension
purchase-history Spending analysis by store, date range, or wine
get-wishlist Your wishlist with ratings, notes, and prices
consumption-history Wines you've opened — by name, color, or date range
tasting-notes Your tasting notes and reviews with ratings and scores
refresh-data Force a fresh pull (auto-refreshes every 24 hours)
clear-user-data Remove stored credentials and cached data from this machine (Claude Code plugin only)

Included skills

Note: Skills are available in Cowork and Code modes via the Claude Code plugin. Chat mode (Desktop Extension) provides tools only.

cellartracker-data — Teaches Claude to interpret CellarTracker data: table relationships, score abbreviations, drinking windows, and query routing.

wine-purchase-evaluator — Framework for evaluating wine purchases. Two-score system (Quality + Personal Fit) with BUY/CONSIDER/PASS verdicts. Checks your cellar for redundancy, verifies pricing, and applies drinking window discipline.

To customize the evaluator for your preferences, copy skills/wine-purchase-evaluator/references/preferences-example.md to preferences.md in the same directory and edit it.

How it works

CellarTracker has no official API. This server uses their CSV export endpoint, which authenticates with your username and password over HTTPS. Data is cached locally and auto-refreshes every 24 hours.

The server can only access your data — inventory, purchases, notes, and wishlist. It cannot search CellarTracker's full wine database.

Security & credentials

CellarTracker API limitations

CellarTracker has no OAuth, API keys, or scoped tokens. Authentication requires your actual account username and password, sent as URL query parameters over HTTPS. While encrypted on the wire, query parameters are routinely logged in server-side access logs on CellarTracker's infrastructure. There is no way to create read-only or limited-access credentials, therefore, this MCP only performs read operations, but it authenticates with your full account.

How this server protects your credentials

Protection Details
File permissions Config directory 0700, .env file 0600 — only your OS user can read
Error stripping Network errors are caught and re-thrown without the URL, which contains credentials
No logging Credentials never appear in stdout, stderr, or error messages
OS keychain Desktop Extension (.mcpb) stores credentials in macOS Keychain / Windows Credential Manager
Env var support Set CT_USERNAME / CT_PASSWORD environment variables to avoid storing credentials on disk

The Claude Code plugin stores credentials as plaintext in ~/.config/cellartracker-mcp/.env. We evaluated OS keychain integration (#22) and decided against it. The native dependency cost (node-gyp / keytar) outweighs the security benefit for wine cellar data, and the Desktop Extension path already uses the OS keychain.

Recommendations

  • Use a unique password for CellarTracker. Do not reuse a password from other services.
  • Prefer environment variables over the setup-credentials tool if you want to avoid persisting credentials to disk.
  • Pin to a specific version in your MCP config (e.g., cellartracker-mcp@0.3.0) rather than relying on @latest.

Development

For contributors or anyone who wants to run from source:

git clone https://github.com/slavins-co/cellartracker-mcp.git
cd cellartracker-mcp
npm install
npm run build

# Set credentials
cp .env.example .env
# Edit .env with your CT login

# Test as Claude Code plugin
claude --plugin-dir .

AI Disclosure

This project was developed with AI assistance.

Trademark Disclosure

This project is in no way affiliated or connected to CellarTracker! LLC. "CellarTracker" is a trademark of CellarTracker! LLC.

License

MIT

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
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
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
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
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
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
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
Exa MCP Server

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。

官方
精选