anylist-mcp-server

rosskukulinski/anylist-mcp-server

3.2

If you are the rightful owner of anylist-mcp-server and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to dayong@mcphub.com.

A Model Context Protocol (MCP) server for AnyList, enabling AI-assisted management of recipes, collections, and meal planning.

Anylist MCP Server

A Model Context Protocol (MCP) server for AnyList - manage your recipes, recipe collections, and meal planning through AI assistants.

Built using the unofficial anylist SDK.

Features

Recipe Management

  • list_recipes - Get all recipes, optionally filtered by collection
  • get_recipe - Get full recipe details including ingredients, prep/cook times
  • search_recipes - Search recipes by name or ingredient
  • create_recipe - Create a new recipe with ingredients, steps, and timing
  • update_recipe - Modify an existing recipe
  • delete_recipe - Remove a recipe

Recipe Collections

  • list_recipe_collections - Get all recipe collections
  • create_recipe_collection - Create a new collection
  • add_recipe_to_collection - Add a recipe to a collection
  • remove_recipe_from_collection - Remove a recipe from a collection
  • delete_recipe_collection - Delete a collection

Meal Planning

  • create_meal_plan - Schedule a recipe on the meal planning calendar with meal type (breakfast, lunch, dinner, snack)
  • get_meal_plan - Get planned meals for a date range

Installation

npm

npm install -g anylist-mcp-server

Docker

docker pull rosskukulinski/anylist-mcp-server

Configuration

The server requires your AnyList credentials via environment variables:

export ANYLIST_EMAIL="your-email@example.com"
export ANYLIST_PASSWORD="your-password"

Optionally, you can specify a custom credentials file location:

export ANYLIST_CREDENTIALS_FILE="/path/to/credentials"

Usage with Claude Desktop

Add the following to your Claude Desktop configuration file:

macOS

~/Library/Application Support/Claude/claude_desktop_config.json

Windows

%APPDATA%\Claude\claude_desktop_config.json

npm installation

{
  "mcpServers": {
    "anylist": {
      "command": "anylist-mcp-server",
      "env": {
        "ANYLIST_EMAIL": "your-email@example.com",
        "ANYLIST_PASSWORD": "your-password"
      }
    }
  }
}

Docker installation

{
  "mcpServers": {
    "anylist": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e", "ANYLIST_EMAIL",
        "-e", "ANYLIST_PASSWORD",
        "rosskukulinski/anylist-mcp-server"
      ],
      "env": {
        "ANYLIST_EMAIL": "your-email@example.com",
        "ANYLIST_PASSWORD": "your-password"
      }
    }
  }
}

Example Prompts

Once configured, you can ask your AI assistant things like:

  • "What recipes do I have?"
  • "Show me the ingredients for my pasta recipe"
  • "Search for recipes with chicken"
  • "Create a new recipe for chocolate chip cookies with these ingredients..."
  • "Add my pasta recipe to the 'Weeknight Dinners' collection"
  • "Plan dinner for tomorrow with the chicken stir fry recipe"
  • "What meals do I have planned for this week?"

Development

# Clone the repository
git clone https://github.com/rosskukulinski/anylist-mcp-server.git
cd anylist-mcp-server

# Install dependencies
npm install

# Build
npm run build

# Run locally
ANYLIST_EMAIL="your-email" ANYLIST_PASSWORD="your-password" npm start

License

MIT License - see for details.

Disclaimer

This project uses an unofficial, reverse-engineered API wrapper for AnyList. It is not affiliated with or endorsed by AnyList. Use at your own risk.