Skip to main content
Glama

Tandoor MCP Server

A FastMCP server for interacting with Tandoor Recipe Manager via the Model Context Protocol.

Features

  • Recipe Management: Search, view, create, and import recipes from URLs

  • Cooking History: Track what you've cooked with ratings and notes

  • Recipe Suggestions: Get recipe ideas based on pantry inventory

  • Meal Planning: Plan meals with bulk scheduling support

  • Shopping Lists: Manage shopping with bulk add/remove/check operations

  • Pantry Tracking: Track on-hand ingredients for smarter suggestions

  • Lookups: Search keywords, foods, and units

Tools (4 Consolidated Tools)

tandoor_recipes

Operation

Description

Bulk Support

list

Search recipes by name, keywords, foods, or rating

-

get

Get full recipe with ingredients and instructions

-

create

Create a new recipe

-

import

Import recipe from URL

urls list

history

Get cooking history with date filtering

-

log

Log a cooked recipe with rating/notes

logs list

suggest

Get recipe suggestions based on pantry

-

tandoor_meal_plan

Operation

Description

Bulk Support

types

List available meal types

-

list

Get meal plan entries with date filtering

-

create

Add recipe to meal plan

meals list

create_type

Create a new meal type

-

delete

Delete a meal plan entry

meal_plan_ids list

tandoor_shopping_list

Operation

Description

Bulk Support

list

Get shopping list items

-

add

Add item to shopping list

add_items list

update

Update a shopping list item

-

remove

Remove item from shopping list

item_ids list

check

Check/uncheck items

item_ids or items list

clear

Clear checked items and sync to pantry

-

tandoor_lookup

Type

Description

Bulk Support

keywords

Search recipe keywords/tags

-

foods

Search foods/ingredients

-

units

Search measurement units

-

pantry

View/update pantry inventory

pantry_items list

Bulk Operations Examples

Plan a week of meals

tandoor_meal_plan(
    operation="create",
    meals=[
        {"recipe": "Pasta Carbonara", "date": "2025-01-27", "meal_type": 3, "servings": 4},
        {"recipe": "Chicken Curry", "date": "2025-01-28", "meal_type": 3, "servings": 4},
        {"recipe": "Beef Tacos", "date": "2025-01-29", "meal_type": 3, "servings": 4},
    ]
)

Add multiple items to shopping list

tandoor_shopping_list(
    operation="add",
    add_items=[
        {"food": "Milk", "amount": 1, "unit": "gallon"},
        {"food": "Eggs", "amount": 12},
        {"food": "Bread", "amount": 1, "unit": "loaf"},
    ]
)

Update pantry inventory

tandoor_lookup(
    type="pantry",
    action="update",
    pantry_items=[
        {"food_name": "salt", "on_hand": True},
        {"food_name": "pepper", "on_hand": True},
        {"food_id": 123, "on_hand": False},
    ]
)

Log multiple cooked recipes

tandoor_recipes(
    operation="log",
    logs=[
        {"recipe_id": "45", "servings": 4, "rating": 5, "comment": "Perfect!"},
        {"recipe_id": "32", "servings": 2, "rating": 4},
    ]
)

Configuration

Set the following environment variables:

Variable

Description

TANDOOR_URL

Tandoor instance URL (default: http://web_recipes)

TANDOOR_PUBLIC_URL

Public URL for recipe links (optional)

TANDOOR_API_TOKEN

API token from Tandoor settings

Running with Docker

services:
  mcp_server:
    build: .
    environment:
      - TANDOOR_URL=http://your-tandoor-instance
      - TANDOOR_PUBLIC_URL=https://recipes.example.com
      - TANDOOR_API_TOKEN=your_token_here
    ports:
      - "8089:8000"

Running Locally

pip install -r requirements.txt
export TANDOOR_URL=http://localhost:8081
export TANDOOR_API_TOKEN=your_token_here
python server.py

MCP Client Configuration

{
  "mcpServers": {
    "tandoor": {
      "url": "http://localhost:8089/mcp/"
    }
  }
}

License

MIT

-
security - not tested
F
license - not found
-
quality - not tested

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/oculairmedia/Tandoor.mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server