weekplan-mcp-server
MCP server enabling AI assistants to manage recipes and ingredients in the WeekPlan app via its REST API.
README
WeekPlan MCP Server
An MCP (Model Context Protocol) server that gives AI assistants like Claude the ability to manage recipes and ingredients in your WeekPlan app via its REST API. See WeekPlan for the main app. See WeekPlan Watch for an Apple Watch companion app.
What It Does
Once connected, Claude can:
- List, add, and delete ingredients (name + unit)
- List, add, and delete recipes (name, ingredients with quantities, and step-by-step instructions)
- Automatically create missing ingredients when adding a recipe
Prerequisites
- Node.js 18+ — Download here
- A running WeekPlan API server (defaults to
http://localhost:3000)
Installation
Option 1 — Clone and build (recommended for local development)
git clone <your-repo-url> weekplan-mcp-server
cd weekplan-mcp-server
npm install
npm run build
The compiled server will be at dist/index.js.
Option 2 — Install globally from npm (once published)
npm install -g weekplan-mcp-server
This makes the weekplan-mcp-server command available system-wide.
Configuration
The server is configured via a single environment variable:
| Variable | Default | Description |
|---|---|---|
WEEKPLAN_URL |
http://localhost:3000 |
Base URL of your WeekPlan REST API |
Adding to Claude Code
Run the following command to register the server with Claude Code:
claude mcp add weekplan-mcp-server -e WEEKPLAN_URL=http://localhost:3000 -- node /absolute/path/to/weekplan-mcp-server/dist/index.js
Replace /absolute/path/to/weekplan-mcp-server with the actual path where you cloned the repo.
Or edit ~/.claude.json to add it to a specific project.
Or edit claude_desktop_config.json directly (see the Claude Desktop section below — the format is the same).
Adding to Claude Desktop
Edit your Claude Desktop config file:
- macOS:
~/Library/'Application Support'/Claude/claude_desktop_config.json - Windows:
%APPDATA%\Claude\claude_desktop_config.json - Linux:
~/.config/Claude/claude_desktop_config.json
Add the server under mcpServers:
{
"mcpServers": {
"weekplan": {
"command": "node",
"args": ["/absolute/path/to/weekplan-mcp-server/dist/index.js"],
"env": {
"WEEKPLAN_URL": "http://localhost:3000"
}
}
}
}
Restart Claude Desktop after saving the file.
Adding to Cursor
Open Cursor settings and navigate to Features → MCP Servers, then add:
{
"weekplan": {
"command": "node",
"args": ["/absolute/path/to/weekplan-mcp-server/dist/index.js"],
"env": {
"WEEKPLAN_URL": "http://localhost:3000"
}
}
}
Adding to VS Code (Copilot / MCP extension)
In your .vscode/mcp.json or user settings:
{
"servers": {
"weekplan": {
"type": "stdio",
"command": "node",
"args": ["/absolute/path/to/weekplan-mcp-server/dist/index.js"],
"env": {
"WEEKPLAN_URL": "http://localhost:3000"
}
}
}
}
Available Tools
Ingredients
| Tool | Description | Parameters |
|---|---|---|
weekplan_list_ingredients |
List all ingredients | (none) |
weekplan_add_ingredient |
Add or update an ingredient | name (string), unit (string) |
weekplan_delete_ingredient |
Delete an ingredient | id (string) |
Recipes
| Tool | Description | Parameters |
|---|---|---|
weekplan_list_recipes |
List all recipes | (none) |
weekplan_add_recipe |
Add or update a recipe | name (string), ingredients (array), steps (array of strings) |
weekplan_delete_recipe |
Delete a recipe | id (string) |
The ingredients array in weekplan_add_recipe accepts objects with:
ingredientName— name of the ingredient (created automatically if missing)quantity— numeric amountunit— unit of measurement (e.g."g","ml","tbsp")
IDs are generated automatically from names (e.g. "Olive Oil" → "olive-oil"), so add operations are idempotent — running them twice won't create duplicates.
Example Prompts
Once the server is connected, try asking Claude:
Add a recipe for spaghetti bolognese with ingredients and steps.
List all my ingredients.
Add 500g of chicken breast as an ingredient.
Delete the recipe for lasagne.
Development
# Run directly from TypeScript (no build step needed)
npm run dev
# Build to dist/
npm run build
# Run the compiled server
npm start
The server communicates over stdio, so it has no open ports and is safe to run as a subprocess managed by your AI client.
Project Structure
src/
├── index.ts # Entry point — connects stdio transport
├── server.ts # Registers all tools with the MCP server
├── api-client.ts # HTTP client for the WeekPlan REST API
├── slugify.ts # Converts names to stable IDs
└── tools/
├── ingredients.ts # Ingredient tool handlers
└── recipes.ts # Recipe tool handlers
Troubleshooting
"Cannot connect to WeekPlan API"
Make sure your WeekPlan server is running and WEEKPLAN_URL points to the correct address.
"command not found: node" Node.js is not installed or not on your PATH. Install it from nodejs.org.
Tools don't appear in Claude
- Confirm the path in your config points to
dist/index.js(notsrc/index.ts) - Make sure you ran
npm run buildafter cloning - Restart your AI client after changing the config
Build errors
Run npm install first to ensure all dependencies are present, then npm run build.
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。