mealie-mcp-ts

mealie-mcp-ts

Enables AI assistants to manage recipes, meal plans, and shopping lists in Mealie through natural language, supporting CRUD operations and URL-based recipe imports.

Category
访问服务器

README

Mealie MCP Server

An MCP (Model Context Protocol) server that enables LLMs to interact with Mealie - a self-hosted recipe manager and meal planner.

Features

  • Recipe Management: List, view, create, update, and delete recipes
  • Recipe Import: Import recipes from URLs (supports most recipe websites)
  • Ingredient Parsing: Automatic parsing of natural language ingredients (e.g., "2 cups flour")
  • Image Uploads: Upload images to recipes via base64 encoding
  • Meal Planning: Create and manage meal plans with date scheduling
  • Shopping Lists: Full CRUD for shopping list items, plus add recipe ingredients to lists
  • Organization: Manage categories and tags for recipe organization
  • Food Management: List and create food items

Installation

# Clone the repository
git clone https://github.com/your-username/mealie-mcp.git
cd mealie-mcp

# Install dependencies
pnpm install

# Build
pnpm run build

Configuration

Create a .env file based on .env.example:

cp .env.example .env

Required environment variables:

Variable Description
MEALIE_URL URL of your Mealie instance (e.g., http://localhost:9925)
MEALIE_API_KEY API key from Mealie (Settings → API Tokens)
ENABLED_TOOLS (Optional) Comma-separated list of tools to enable

Usage

With Claude Desktop

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

{
  "mcpServers": {
    "mealie": {
      "command": "node",
      "args": ["/path/to/mealie-mcp/dist/index.js"],
      "env": {
        "MEALIE_URL": "http://localhost:9925",
        "MEALIE_API_KEY": "your-api-key"
      }
    }
  }
}

Remote HTTP mode (e.g. on a NAS)

For running the server on a different machine (Synology NAS, home server, etc.), use the HTTP transport. The SDK's Streamable HTTP transport lets Claude Desktop connect over the network.

Set these env vars:

MCP_TRANSPORT=http
PORT=3000
ALLOWED_HOSTS=nas.local,192.168.1.50   # Host header allowlist (DNS rebind protection)
MEALIE_URL=http://mealie:9000           # service name if on same compose network
MEALIE_API_KEY=...

Claude Desktop config:

{
  "mcpServers": {
    "mealie": {
      "url": "http://nas.local:3000/mcp"
    }
  }
}

With Docker

The bundled Dockerfile defaults to HTTP transport on port 3000.

docker build -t mealie-mcp .

docker run -p 3000:3000 \
  -e MEALIE_URL=http://host.docker.internal:9925 \
  -e MEALIE_API_KEY=your-api-key \
  -e ALLOWED_HOSTS=localhost \
  mealie-mcp

Compose service (attach to existing Mealie stack)

  mealie-mcp:
    build: ./mealie-mcp
    container_name: mealie-mcp
    restart: unless-stopped
    ports:
      - "3000:3000"
    environment:
      - MEALIE_URL=http://mealie:9000
      - MEALIE_API_KEY=${MEALIE_API_KEY}
      - ALLOWED_HOSTS=nas.local,192.168.1.50
    depends_on:
      - mealie

Available Tools (25 total)

Recipes (6 tools)

Tool Description
list_recipes List all recipes with pagination
get_recipe Get full recipe details by slug
create_recipe Create a new recipe with ingredients and instructions
update_recipe Update an existing recipe
delete_recipe Delete a recipe
upload_recipe_image Upload an image to a recipe (base64)

Recipe Import (2 tools)

Tool Description
test_scrape_url Test if a URL can be scraped for recipe data
create_recipe_from_url Import a recipe by scraping a URL

Meal Planning (4 tools)

Tool Description
list_meal_plans List meal plans with optional date range filter
get_todays_meal_plan Get today's planned meals
create_meal_plan Create a meal plan entry for a specific date
delete_meal_plan Delete a meal plan entry

Shopping Lists (6 tools)

Tool Description
list_shopping_lists List all shopping lists
get_shopping_list Get shopping list with items
add_shopping_list_item Add an item to a shopping list
update_shopping_list_item Update an item (quantity, note, or check/uncheck)
delete_shopping_list_item Remove an item from a shopping list
add_recipe_to_shopping_list Add all ingredients from a recipe to a list

Organization (4 tools)

Tool Description
list_categories List all recipe categories
create_category Create a new category
list_tags List all recipe tags
create_tag Create a new tag

Foods (2 tools)

Tool Description
list_foods List all foods/ingredients
create_food Create a new food item

Debugging (1 tool)

Tool Description
get_current_user Get current API user info including group and household context

Important: Group Context

Mealie scopes data by group and household. When you create an API token, it inherits the group context of your current session. If recipes or other items you create via the API aren't visible in your browser:

  1. Use the get_current_user tool to see which group the API token is associated with
  2. Ensure your browser session is viewing the same group (check the URL path)
  3. If needed, switch groups in Mealie's UI, then generate a new API token

Development

# Run in development mode
pnpm run dev

# Lint
pnpm run lint

# Format
pnpm run format

# Build
pnpm run build

Local Mealie Instance

Start a local Mealie instance for testing:

docker-compose up -d

This starts Mealie at http://localhost:9925.

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