Scraps Kitchen MCP Server

Your AI assistant's cooking brain. Scraps Kitchen gives any AI agent persistent, household-aware kitchen intelligence — pantry tracking, dietary safety, recipe management, shopping lists, and more.

Unlike generic chatbot memory, Scraps maintains structured cooking data: what's in your fridge, who you cook for, their allergies and preferences, your recipe history with cook notes, and your kitchen equipment. Any MCP-compatible agent can tap into this to have genuinely personalized cooking conversations.

Live server: https://api.scraps.kitchen/mcp (Streamable HTTP) Web app: scraps.kitchen

Quick Start

Claude Desktop

Add to your claude_desktop_config.json:

{
  "mcpServers": {
    "scraps-kitchen": {
      "type": "streamable-http",
      "url": "https://api.scraps.kitchen/mcp"
    }
  }
}

VS Code / Cursor

Add to your MCP settings:

{
  "servers": {
    "scraps-kitchen": {
      "type": "streamable-http",
      "url": "https://api.scraps.kitchen/mcp"
    }
  }
}

On first use, you'll be prompted to authenticate via OAuth. Sign up at scraps.kitchen if you don't have an account. One free tool (suggest_quick_meal) works without authentication.

Tools (27)

Meal Suggestions

Tool	Auth	Description
`suggest_quick_meal`	No	Generate a recipe from ingredients. 1 free call per IP per 24 hours.

Pantry (6 tools)

Tool	Description
`get_pantry`	List pantry items with category, quantity, stock status, perishability, and storage hints. Filter by in-stock or stale items.
`add_pantry_items`	Add up to 50 items at once. Auto-enriched with food intelligence.
`update_pantry_item`	Update name, quantity, unit, category, or stock status.
`remove_pantry_item`	Permanently delete a pantry item.
`mark_ingredients_used`	Mark ingredients as consumed after cooking. Fuzzy name matching. Staples auto-skipped.
`confirm_pantry_fresh`	Reset staleness clock on items the user confirms are still good.

Shopping List (5 tools)

Tool	Description
`get_shopping_list`	List shopping items with checked status. Filter by bought/remaining.
`add_shopping_list_item`	Add a single item to the list.
`add_recipe_to_shopping_list`	Add all ingredients from a saved recipe at once.
`check_off_shopping_item`	Mark items as bought or still needed.
`remove_shopping_list_item`	Remove a single item.
`clear_checked_shopping_items`	Remove all checked-off items at once (post-shopping cleanup).

Household (4 tools)

Tool	Description
`get_household`	List household members with allergens, restrictions, preferences, dislikes, goals, and life stages.
`add_diner`	Add a household member with dietary profile.
`update_diner`	Update dietary details. Array fields replace entirely — send the full list.
`remove_diner`	Remove a member and their recipe ratings.

Equipment (4 tools)

Tool	Description
`get_equipment`	List kitchen equipment (tools, appliances).
`add_equipment`	Add a piece of equipment with optional notes.
`update_equipment`	Update equipment name or notes.
`remove_equipment`	Remove equipment from inventory.

Recipes (6 tools)

Tool	Description
`get_recipes`	List saved recipes. Search by keyword, filter by status or favorites.
`get_recipe`	Full recipe detail: ingredients, steps, cook notes, diner ratings.
`update_recipe`	Update title, status, rating, favorite, or sharing status.
`delete_recipe`	Permanently delete a recipe and all associated data.
`add_recipe_note`	Record cook notes — observations, modifications, per-cook ratings.
`rate_recipe_for_diner`	Track which household members loved which recipes.

Authentication

Scraps uses OAuth 2.1 with PKCE. Most MCP clients handle this automatically — you'll see a login prompt on first connection.

Scopes:

Scope	Access
`read:pantry`	View pantry inventory
`read:household`	View household dietary info
`read:recipes`	View saved recipes
`write:pantry`	Update pantry
`write:recipes`	Save recipes and cooking notes
`write:shopping`	Manage shopping list
`write:household`	Manage household members
`write:equipment`	Manage kitchen equipment

Discovery: https://api.scraps.kitchen/.well-known/mcp.json

How Agents Use Scraps

Scraps is a data layer, not a reasoning engine. It stores structured kitchen intelligence that makes any AI assistant better at cooking conversations:

Read the kitchen context — get_pantry, get_household, get_equipment, get_recipes
Use your own LLM to reason about what to cook, considering allergies, preferences, and available ingredients
Write back results — mark_ingredients_used, add_recipe_note, rate_recipe_for_diner

Every interaction makes the data richer. More cook history, more diner feedback, more pantry accuracy — which makes the next conversation better.

Example Workflows

"What should I cook tonight?"

get_pantry → see available ingredients
get_household → check dietary constraints and allergens
get_equipment → know what tools are available
get_recipes(status: "cooked") → see what's worked before
Agent reasons about a meal suggestion using all this context

After cooking:

mark_ingredients_used(["chicken thighs", "ginger", "soy sauce"]) → update pantry
add_recipe_note(recipe_id, "Added extra garlic, seared 2 min longer") → build cook history
rate_recipe_for_diner(recipe_id, diner_id, { loved_it: true }) → track preferences

Grocery planning:

get_pantry(stale_only: true) → items past freshness window
get_shopping_list → existing list
add_shopping_list_item("eggs") → add what's needed
After shopping: add_pantry_items([...]) → restock, clear_checked_shopping_items → clean up list

License

MIT

scraps-kitchen-mcp