Tandoor MCP Server
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Tandoor MCP Server create a recipe for chocolate chip cookies"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Tandoor MCP Server
A Model Context Protocol (MCP) server for interacting with Tandoor Recipe Manager.
Current Status
✅ create_tandoor_recipe: Successfully implemented and tested
✅ create_tandoor_meal_plan: Successfully implemented and tested
✅ get_recipes: Successfully implemented and tested
✅ get_meal_plans: Successfully implemented and tested
✅ get_recipe_details: Implemented
✅ get_meal_types: Implemented
✅ get_keywords: Implemented
✅ get_foods: Implemented
✅ get_units: Implemented
✅ get_shopping_list: Implemented
✅ add_shopping_list_item: Implemented (with name lookup)
✅ update_shopping_list_item: Implemented
✅ remove_shopping_list_item: Implemented
Related MCP server: Unofficial AnyList MCP Server
Features
Create recipes in Tandoor with ingredients and instructions
Add recipes to meal plans for specific dates and meal types
Search for recipes using various criteria (name, keywords, foods, rating)
Retrieve meal plans filtered by date range and meal type
Retrieve full details for a specific recipe
List available meal types, keywords, foods, and units
Manage shopping list items (view, add, update, remove)
Setup
Install dependencies:
npm installBuild the server:
npm run buildRun the server manually (for testing):
# Windows PowerShell $env:TANDOOR_URL = "https://your-tandoor-instance.com" $env:TANDOOR_API_TOKEN = "your-api-token" node .\build\index.js # Windows CMD set TANDOOR_URL=https://your-tandoor-instance.com set TANDOOR_API_TOKEN=your-api-token node .\build\index.js # Linux/macOS TANDOOR_URL=https://your-tandoor-instance.com TANDOOR_API_TOKEN=your-api-token node ./build/index.jsConfigure in MCP settings:
{ "mcpServers": { "tandoor": { "command": "node", "args": ["path/to/tandoor-mcp-server/build/index.js"], "env": { "TANDOOR_URL": "https://your-tandoor-instance.com", "TANDOOR_API_TOKEN": "your-api-token" }, "disabled": false, "autoApprove": [] } } }If you want everything autoApprove...
"autoApprove": [
"add_shopping_list_item",
"create_tandoor_meal_plan",
"create_tandoor_recipe",
"get_foods",
"get_keywords",
"get_meal_plans",
"get_meal_types",
"get_recipe_details",
"get_recipes",
"get_recipes",
"get_shopping_list",
"get_units",
"remove_shopping_list_item",
"update_shopping_list_item"
]
Add
./.clinerulesto your project directory
optionally also add
./memory-bankwith a markdown or text file list of foods you like or dislike in case you'd like suggestions for meals from the assistant.
Available Tools
create_tandoor_recipe
Creates a new recipe in Tandoor.
Parameters:
name(string, required): The name of the recipe.description(string, optional): Description for the recipe.servings(number, optional): Number of servings.ingredients_block(string, required): A multi-line block of text listing ingredients, one per line.instructions_block(string, required): A multi-line block of text detailing the recipe instructions.
Example:
{
"name": "Simple Pasta",
"description": "A quick and easy pasta dish.",
"servings": 2,
"ingredients_block": "200g pasta\n2 tbsp olive oil\n1 clove garlic, minced\nSalt and pepper to taste",
"instructions_block": "1. Cook pasta according to package instructions.\n2. Heat oil in a pan and add garlic.\n3. Drain pasta and add to the pan.\n4. Season with salt and pepper."
}create_tandoor_meal_plan
Adds one or more recipes to the Tandoor meal plan for a specific date and meal type.
Parameters:
title(string, optional): Title for the meal plan entry.recipes(array, required): Array of recipe names or IDs to add to the plan.start_date(string, required): The date for the meal plan entry (YYYY-MM-DD).meal_type(string, required): The name of the meal type (e.g., 'Dinner', 'Lunch').servings(number, optional): Number of servings for the meal plan entry (default: 1).note(string, optional): Note for the meal plan entry.
Example:
{
"title": "Friday Dinner",
"recipes": [123, "Simple Pasta"],
"start_date": "2025-03-29",
"meal_type": "Dinner",
"servings": 2,
"note": "Quick dinner for Friday night."
}get_recipes
Search for recipes in Tandoor based on various criteria.
Parameters:
query(string, optional): Search term to match against recipe names (fuzzy match).keywords(array of integers, optional): Array of Keyword IDs. Returns recipes matching ANY of these keywords.foods(array of integers, optional): Array of Food IDs. Returns recipes containing ANY of these foods.rating(integer, optional): Minimum rating (0-5) the recipe should have.limit(integer, optional): Maximum number of recipes to return. Defaults to 10.
Example:
{
"query": "chicken",
"keywords": [5, 12],
"rating": 4,
"limit": 5
}get_meal_plans
Retrieve meal plan entries from Tandoor, optionally filtering by date range and meal type.
Parameters:
from_date(string, optional): Start date (YYYY-MM-DD) to filter meal plans (inclusive).to_date(string, optional): End date (YYYY-MM-DD) to filter meal plans (inclusive).meal_type_id(integer, optional): Meal Type ID to filter by.
Example:
{
"from_date": "2025-03-01",
"to_date": "2025-03-31",
"meal_type_id": 1
}get_recipe_details
Retrieve the full details of a specific recipe.
Parameters:
recipe_id(integer, required): The ID of the recipe to retrieve.
Example:
{
"recipe_id": 123
}get_meal_types
List all available meal types in Tandoor.
Parameters: None
Example:
{}get_keywords
List or search for keywords.
Parameters:
query(string, optional): Optional search term for keyword name.root(integer, optional): Optional ID to get first-level children (0 for root).tree(integer, optional): Optional ID to get all children in a tree.
Example:
{
"query": "italian"
}get_foods
List or search for foods.
Parameters:
query(string, optional): Optional search term for food name.root(integer, optional): Optional ID to get first-level children (0 for root).tree(integer, optional): Optional ID to get all children in a tree.
Example:
{
"query": "chicken breast"
}get_units
List or search for units.
Parameters:
query(string, optional): Optional search term for unit name.
Example:
{
"query": "gram"
}get_shopping_list
Retrieve the current shopping list items.
Parameters:
checked(string, optional): Filter by checked status ("true", "false", "both", "recent"). Defaults to "recent".
Example:
{
"checked": "false"
}add_shopping_list_item
Add an item to the shopping list, allowing food/unit names or IDs.
Parameters:
food_name_or_id(string or integer, required): The name or ID of the food item.amount(string, required): The amount needed (e.g., '1', '2.5', '1/2').unit_name_or_id(string or integer, required): The name or ID of the unit (e.g., 'cup', 'g', 5).note(string, optional): Optional note for the item.
Example (using names):
{
"food_name_or_id": "Chicken Breast",
"amount": "500",
"unit_name_or_id": "g",
"note": "For stir-fry"
}Example (using IDs):
{
"food_name_or_id": 42,
"amount": "2",
"unit_name_or_id": 15
}update_shopping_list_item
Update an existing shopping list item (e.g., check/uncheck, change amount).
Parameters:
item_id(integer, required): The ID of the shopping list item to update.amount(string, optional): Optional new amount.unit_id(integer, optional): Optional new unit ID.checked(boolean, optional): Optional new checked status.note(string, optional): Optional new note.
Example (checking off an item):
{
"item_id": 101,
"checked": true
}remove_shopping_list_item
Remove an item from the shopping list.
Parameters:
item_id(integer, required): The ID of the shopping list item to remove.
Example:
{
"item_id": 102
}Testing
A test script is included to help verify the server functionality without relying on the MCP connection:
# Set your API token (if not already set in the environment)
$env:TANDOOR_API_TOKEN = "your-api-token"
# Run the test script
npm run testThe test script provides a simple menu to:
List available tools
Test creating a recipe
Test adding a recipe to a meal plan
Implementation Notes
The Tandoor API has some specific requirements for creating meal plans:
The
from_datefield must include a time component (YYYY-MM-DDT00:00:00)The
recipefield must be an object withid,name, andkeywordspropertiesThe
meal_typefield must be an object withidandnamepropertiesThe
servingsfield must be a string, not a number
These requirements are handled automatically by the MCP server.
Troubleshooting
If you encounter issues with the MCP server:
Ensure the server is running and showing
[Setup] Tandoor MCP server running on stdio.Check that your Tandoor URL and API token are correct
Look for error messages in the terminal where the server is running
Verify that the meal type names match exactly what's in your Tandoor instance
Use the test script to verify server functionality outside of the MCP system
Try running the server manually and check the console output for detailed error messages
Available Tools
13 toolsadd_shopping_list_itemB
Add an item to the shopping list, allowing food/unit names or IDs.
Inputs: amount (required), food_name_or_id (required), unit_name_or_id (required), note.
create_tandoor_meal_planA
Add one or more recipes to the Tandoor meal plan for a specific date and meal type.
Inputs: recipes (required), meal_type (required), start_date (required), note, +2 more.
create_tandoor_recipeB
Create a new recipe in Tandoor.
Inputs: name (required), ingredients_block (required), instructions_block (required), servings, +1 more.
get_foodsC
List or search for foods.
Inputs: root, tree, query.
get_keywordsC
List or search for keywords.
Inputs: root, tree, query.
get_meal_plansA
Retrieve meal plan entries from Tandoor, optionally filtering by date range and meal type.
Inputs: to_date, from_date, meal_type_id.
get_meal_typesA
List all available meal types in Tandoor.
Inputs: no named parameters.
get_recipe_detailsA
Retrieve the full details of a specific recipe.
Inputs: recipe_id (required).
get_recipesC
Search for recipes in Tandoor based on various criteria.
Inputs: foods, limit, query, rating, +1 more.
get_shopping_listA
Retrieve the current shopping list items.
Inputs: checked.
get_unitsB
List or search for units.
Inputs: query.
remove_shopping_list_itemB
Remove an item from the shopping list.
Inputs: item_id (required).
update_shopping_list_itemA
Update an existing shopping list item (e.g., check/uncheck, change amount).
Inputs: item_id (required), note, amount, checked, +1 more.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 13 tools
Most tools target distinct resources and actions. The only slight overlap is between get_recipe_details (fetch one recipe) and get_recipes (search), but their descriptions clearly differentiate them.
The tool names mostly follow a verb_noun pattern, but there are inconsistencies: 'create_tandoor_recipe' and 'create_tandoor_meal_plan' use 'tandoor' while others do not, and the shopping list tools use 'add/update/remove' rather than a consistent 'create/update/delete'. Singular/plural usage also differs (get_recipe_details vs get_recipes).
13 tools is well within the ideal 3-15 range and each tool serves a clear purpose for managing recipes, meal plans, and shopping lists.
The shopping list has full CRUD coverage, but recipes and meal plans only support creation and retrieval, with no update or delete operations. This is a significant gap that agents will likely encounter when trying to manage or correct existing recipes and meal plans.
Maintenance
Related MCP Connectors
Household-aware cooking brain: pantry, meal suggestions, dietary safety, recipes, shopping lists.
Official Todoist MCP server. Manage tasks and projects. Hosted, no API key needed.
MCP server for Google Tasks: review, plan and tidy your tasks by asking. Deletes always ask first.
AnyList MCP server for ChatGPT, Claude and any AI agent: list shopping lists, add list item, cross off item, read recipe, create recipe, add recipe to meal plan and add meal plan ingredients to list in your own AnyList account. Cotlet works in the account through your Cloud Browser, a real Chrome where you sign in to AnyList yourself; Cotlet does not store your password and you need no AnyList API key. About the official AnyList API: AnyList has no public API, developer docs or official MCP server (checked 25 Oct 2026). The official connections are voice entry with Siri and Amazon Alexa, and an iCalendar link that shows your meal plan in a calendar app. Setup and FAQ: https://cotlet.dev/integrations/anylist-api-mcp-server
Related MCP Servers
- AlicenseBqualityCmaintenanceMCP server for Mealie that exposes its REST API to manage recipes, meal plans, shopping lists, cookbooks, and taxonomy through natural language.7558 PyPIMIT
- AlicenseNot gradedqualityAmaintenanceMCP server that integrates with AnyList for managing shopping lists, recipes, and meal planning via natural language.35ISC
- AlicenseNot gradedqualityDmaintenanceMCP server for MealMastery AI meal planning that enables users to manage meal plans, recipes, and grocery lists through natural language conversation with AI agents like Claude.52 npmMIT
- FlicenseAqualityDmaintenanceMCP server for Tandoor Recipes, enabling recipe management (create, edit, search, delete) and entity operations (foods, units, keywords) via natural language from Claude.11-