Spoonacular Extended MCP for AI. Plan meals or find recipes from anything you have.
Works with every AI agent you already use
…and any MCP-compatible client








How this MCP server connects to your AI agent
Spoonacular Extended lets your AI client search thousands of recipes by cuisine, specific ingredients, or exact nutritional metrics. You can find meals that fit dietary needs—like low-carb or high protein—and even pull structured recipe data from any random website URL.
What AI agents can do with Spoonacular Automation
Add to meal plan
Adds a specified recipe or item to the user's ongoing meal plan record.
Autocomplete ingredient search
Suggests full ingredient names when you type part of an ingredient name.
Autocomplete recipe search
Offers suggested recipe titles as you type a partial search query.
You list the items in your kitchen, and the agent returns viable recipes that maximize the use of those specific ingredients.
The agent filters recipe databases to find options meeting strict macro requirements (e.g., 30g protein, under 500 calories).
You provide a link to a cooking blog post, and the agent pulls structured data—like ingredient lists and steps—out of the article.
The system creates multi-day dining schedules based on user inputs or random generation, providing structure to your diet.
You tell the agent a nutritional target (e.g., 'I need 15g of fiber'), and it computes how much of an ingredient is needed to hit that mark.
Ask an AI about this
Waiting for input…
What AI agents can do with Spoonacular Extended MCP Server: 22 Tools for Food Data Management
Use these 22 tools to manage everything from ingredient lookups and nutritional analysis to complex meal plan generation via your AI client.
Make your AI actually useful.
Add this MCP to Claude, Cursor, or Windsurf and your AI stops guessing. It gets real tools to look things up, take action, and handle the stuff you keep doing by hand.
Start using Spoonacular on VinkiusAdd To Meal Plan
Adds a specified recipe or item to the user's ongoing meal plan record.
Autocomplete Ingredient Search
Suggests full ingredient names when you type part of an ingredient name.
Autocomplete Recipe Search
Offers suggested recipe titles as you type a partial search query.
Search Recipes
Searches through thousands of recipes using advanced filters like diet, cuisine, and...
Compute Ingredient Amount
Calculates how much of an ingredient is needed to reach a specific nutritional goal...
Extract Recipe
Pulls structured recipe data and instructions from any provided website URL.
Find Recipes By Ingredients
Returns recipes that prioritize using the list of ingredients you currently have available.
Find Recipes By Nutrients
Searches for recipes based on defined nutritional parameters (e.g., max calories...
Generate Meal Plan
Creates a complete meal plan for the user across specified days.
Get Ingredient Information
Retrieves detailed nutritional and source information about a specific ingredient...
Get Ingredient Substitutes
Provides alternative ingredients that can replace a primary ingredient in a recipe.
Get Meal Plan Week
Retrieves an existing meal plan structured by week for review or modification.
Get Menu Item Information
Gets specific details about a single, named menu item (like from a restaurant).
Get Product Information
Fetches data and nutritional facts for a specific grocery product.
Get Random Recipes
Returns a list of completely random recipes to break menu fatigue.
Get Recipe Information
Pulls the full, detailed instructions and data for one specific recipe ID.
Get Shopping List
Retrieves a user's current shopping list based on past meal plans or recipes.
Get Similar Recipes
Finds other recipes that share similar ingredients, cuisine types, or nutritional profiles with a given recipe ID.
Search Ingredients
Performs a broad search across all available ingredient data.
Search Menu Items
Searches for restaurant menu items using keywords or categories.
Search Products By Upc
Looks up grocery products by scanning or entering their UPC barcode number.
Search Grocery Products
Searches the database for common grocery products by name or category.
Security and governance baked right in.
Pick your AI client below to get set up. Just create a Vinkius account, subscribe, and you're instantly up and running. We handle the entire backend infrastructure, delivering out-of-the-box support for HTTPS Streamable, SSE, and OAuth2—zero messy routing required.
Choose How to Get Started
Build a custom MCP for your own tools, or connect a ready-made integration from our catalog.
Build Your Own
Turn any API into an MCP. Import a spec, define Agent Skills, or deploy with MCPFusion.
- Import from OpenAPI, Swagger, or YAML specs
- Create Agent Skills with progressive disclosure
- Deploy to edge with MCPFusion framework
- Built in DLP, auth, and compliance on every call
- Real time usage dashboard and cost metering
- Publish to catalog or keep private
Make Your AI Do More
Start with Spoonacular, then connect any of our 5,100+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 5,100+ others, all in one place
- Add new capabilities to your AI anytime you want
- Every connection is secured and compliant automatically
- Track usage and costs across all your servers
- Works with Claude, ChatGPT, Cursor, and more
- New servers added to the catalog every week
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Spoonacular. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.
VINKIUS INFRASTRUCTURE
Cloud Hosted
Managed infra
V8 Isolated
Sandboxed per request
Zero-Trust Proxy
No stored credentials
DLP Enforced
Policy on every call
GDPR Compliant
EU data residency
Token Compression
~60% cost reduction
Built on the Model Context Protocol (MCP) for Claude, ChatGPT, Cursor, and more
The Model Context Protocol standardizes how applications expose capabilities to LLMs. Instead of operating in isolation, your AI gains direct access to external platforms, live data, and real-world actions through secure, standardized connections.
This connection provides 22 powerful capabilities that interface natively with Claude, ChatGPT, Cursor, and other compatible AI platforms. No middleware. No custom integration required.
Dealing with recipe blogs shouldn't involve copy-pasting five different sections., Solved with Vinkius AI Gateway
Today, if you want a recipe from an online blog or magazine, you spend time clicking around. You find the ingredients list in one section, but the steps are buried three paragraphs later, formatted as poorly structured text. Then you have to manually copy-paste that mess into your system for processing.
With this MCP server, you just hand the agent the URL. The `extract_recipe` tool does the heavy lifting: it reads the entire page and pulls out only the clean, actionable data—the ingredients list, the measurements, and the step-by-step instructions. It’s structured data, period.
Spoonacular Extended MCP Server gives you precise control over diet planning.
Before this server, planning a week's worth of meals meant opening multiple tabs—one for the meal plan generator, one for nutrient databases, and another for checking ingredient substitutes. It was slow, manual, and prone to calculation errors.
Now you ask your agent: 'Build me a 7-day low-carb menu using only ingredients I have.' The system chains `generate_meal_plan`, checks it with `find_recipes_by_ingredients`, and finishes by running `get_shopping_list`. It handles the whole flow in one go.
What your AI can actually do with this
You connect your AI client to Spoonacular Extended, and it turns into a seriously smart digital sous-chef. You'll be able to find recipes, check nutrition facts, or even pull data from random websites—all without having to click through a dozen tabs. It’s built for when you need accurate results, period.
Finding Recipes Based on What You Have:
You don't gotta waste food just 'cause it's sitting in the back of the fridge. If you tell your agent what ingredients you've got—say, chicken breasts, bell peppers, and rice—it runs find_recipes_by_ingredients, giving you recipes that maximize those specific items. You can also perform a broad search across all available data using search_ingredients or check out detailed information on any single item with get_ingredient_information.
If you're stuck on what to make, it provides alternative ingredients through get_ingredient_substitutes, so you don't have to toss the whole dinner. When you start typing an ingredient name, autocomplete_ingredient_search suggests full names, keeping your search tight.
Hitting Nutritional Targets:
If 'good enough' ain't good enough, the agent handles it. You can use find_recipes_by_nutrients to filter recipes based on strict guidelines—you might need something under 500 calories or with at least 30 grams of protein. If you know your goal is hitting a specific macro count, say 'I gotta get 12g more fiber,' the system runs compute_ingredient_amount, calculating exactly how much of an ingredient you'll need to hit that mark.
You can also search for meals using advanced filters like diet or cuisine via search_recipes.
Planning and Organizing Meals:
The agent helps you plan your whole week, not just tonight’s dinner. Use generate_meal_plan to create a multi-day eating schedule based on what you input. You can then save that plan by running add_to_meal_plan, or review an existing weekly structure with get_meal_plan_week. If you need groceries, it pulls your current shopping list using get_shopping_list.
When you're starving for ideas and don't wanna think, you can ask for random meals via get_random_recipes.
Data Extraction and Research:
It’s not just about recipes. You can give it a link to any cooking blog post using extract_recipe, and it pulls out the structured data—ingredients, steps, everything—so your agent can actually read it. If you find a recipe ID, get_recipe_information gives you every detail. It also helps you find similar meals by running get_similar_recipes, which matches ingredients or profiles to a meal you already like.
For shopping, the system handles both physical stores and restaurant menus. You can search for grocery items generally using search_grocery_products, or check products with their UPC barcode number via search_products_by_upc. If you're out eating, you can use search_menu_items to find specific dishes at a restaurant, and even get details on one item using get_menu_item_information.
You can also look up the facts for any grocery product with get_product_information.
Advanced Searching:
The agent keeps suggestions coming. As you start typing a recipe title, autocomplete_recipe_search offers immediate options. If you just need general info on an ingredient or product, you can use search_products_by_upc, search_menu_items, or the broader search_ingredients. You also get one shot at getting a full inventory of available recipes by using get_similar_recipes.
019e5d58-ae4b-7004-8640-147f30d186fc Here's how it actually works
The bottom line is you can use your AI client to talk to a professional food database that handles the messy search logic for you.
Subscribe to this server and enter your Spoonacular API Key.
Connect the MCP Server to your preferred client (Claude, Cursor, etc.).
Ask your agent a natural language query. For example: 'I have chicken, rice, and broccoli. Give me three recipes using those ingredients.' The agent calls find_recipes_by_ingredients and gives you results.
Who is this actually for?
Dietitians, recipe developers, and serious home cooks. If your job involves tracking specific macros (protein counts, carb limits) or managing complex meal schedules, this is built for you. You're tired of cross-referencing multiple food databases and manually converting recipes from blogs.
Uses extract_recipe to pull structured data from various web sources so it can be used in a single database format.
Runs queries using find_recipes_by_nutrients and compute_ingredient_amount to build client meal plans that hit precise patient macro targets.
Uses the agent to generate random ideas (get_random_recipes) or create weekly menus using generate_meal_plan, saving hours of research time.
What Changes When You Connect
Stop manual searching. Use find_recipes_by_ingredients to instantly generate viable meals based only on what's in your fridge, eliminating 'I don't know what to cook' nights.
Hit precise dietary goals. Instead of guessing, run find_recipes_by_nutrients to guarantee the meal meets specific macro targets (e.g., high fiber, low sodium).
Convert web clutter into data. The extract_recipe tool pulls structured instructions and ingredients from any blog post link so your agent can actually use it.
Manage time better. Use get_shopping_list after running a meal plan to get a consolidated list of everything you need, without having to cross-reference multiple recipes.
Deepen recipe understanding. Need an alternative? Check the ingredient database with get_ingredient_substitutes before writing your grocery list.
See it in action
The Dinner Dilemma
You open the fridge and realize you have chicken, asparagus, and leftover pasta. Instead of wasting time Googling combinations, you ask your agent: 'What can I make with these three things?' The agent runs find_recipes_by_ingredients and immediately gives you a viable recipe using only what you own.
The Macro-Focused Meal
Your client needs to hit 30g of protein while staying under 450 calories. You ask your agent, 'Find me dinner options with at least 30g protein and less than 450 calories.' The agent executes find_recipes_by_nutrients and returns only compliant meals.
The Recipe Blogger
You find a great-looking recipe on a blog, but the steps are messy. You paste the URL and ask your agent to process it. The agent calls extract_recipe and gives you clean, structured data—ingredients list, prep time, instructions—ready for your app.
The Weekly Planner
It's Sunday, and you need a menu for the week. You ask your agent to generate_meal_plan. It sends back seven days of meals. Then, asking it to get_shopping_list compiles every single item needed into one actionable list.
The honest tradeoffs
Searching by vague concepts
Asking the agent: 'Give me a healthy Italian dinner.' This is too broad and requires multiple follow-up questions to get useful results.
Be specific. Use search_recipes with filters, or better yet, give it ingredients. Try: 'Find low-carb Mediterranean recipes using salmon, lemon, and oregano.' The agent uses the precise data in find_recipes_by_ingredients.
Ignoring pantry items
Running a general search like search_recipes when you only have three ingredients. You'll get hundreds of recipes that require things you don't own.
Always start with what you have. Use the find_recipes_by_ingredients tool first. This cuts out 90% of unusable results and focuses on efficiency.
Ignoring nutritional goals
Accepting a recipe because it sounds good, but then realizing it's too high in fat or sodium for the user's diet.
Always check the macros. Run find_recipes_by_nutrients before committing to a meal plan. You can filter by specific values like 'protein > 30g' and 'calories < 500'.
When It Fits, When It Doesn't
Use this MCP Server if your workflow involves multi-step food data processing: reading a recipe from a URL, filtering it against dietary rules, then building a shopping list. You need structured retrieval (e.g., using extract_recipe and find_recipes_by_nutrients). Don't use it if you just need to browse recipes—a standard search engine is faster for that. If your core task is 'What do I have?' or 'What am I missing?', then this tool suite, especially with find_recipes_by_ingredients and get_shopping_list, is essential because it links inventory directly to actionable meal plans.
Questions you might have
How does the find_recipes_by_nutrients tool work? +
It filters recipes against specific macro criteria you set. You tell it, 'Protein must be at least 30g and calories less than 500,' and it only returns matching results.
What is the best tool for planning meals? +
generate_meal_plan creates a full menu schedule. If you need to check what groceries you'll need from that plan, follow up by calling get_shopping_list.
Can I find recipes using only ingredients in my pantry? +
Yes, use the find_recipes_by_ingredients tool. You list your items, and it maximizes usage of what you have on hand.
What if a recipe uses an ingredient I don't know how to spell? +
Use the autocomplete_ingredient_search tool. It provides suggestions as you type, ensuring your input is clean and accurate for the database.
If I get an API key error when running `search_recipes`, what should I check first? +
First, verify that your Spoonacular API Key is correctly entered in the Vinkius connection settings. Second, confirm the key hasn't expired or exceeded its usage quota for the day. The tool won't run if authentication fails.
What should I know about rate limits when calling `generate_meal_plan`? +
The server adheres to standard API rate limits, meaning you can't execute requests too frequently. If you hit the limit, your agent needs a short pause before trying again; it won't automatically retry.
When using `extract_recipe`, what kind of URL works best for conversion? +
The source URL must be directly accessible and contain clear recipe content. URLs pointing to indexes or login screens will fail because the scraper can't find structured data.
How reliable is the information gathered from `search_products_by_upc`? +
The tool pulls data from large product databases, making it highly accurate for common items. However, if the UPC code belongs to a regional or specialty item, you might need to cross-reference the details.
Can I find recipes based on specific ingredients I already have? +
Yes! Use the find_recipes_by_ingredients tool. Just list your ingredients, and the agent will return recipes that maximize the use of what you have while minimizing missing items.
How do I get the full nutritional breakdown of a specific recipe? +
You can use the get_recipe_information tool with the recipe ID and set includeNutrition to true. This provides calories, macros, and vitamins for the dish.
Can the AI extract a recipe from a website link? +
Absolutely. Use the extract_recipe tool and provide the URL. The agent will parse the website and return a structured recipe with ingredients and instructions.
We've already built the connector for Spoonacular Extended. Just plug in your AI agents and start using Vinkius.
No hosting. No infrastructure. No complex setup.
All 22 tools are live and waiting.
You're up and running in seconds.
Vinkius gives your AI agents access to the full catalog of app connectors, all fully managed, secure, and enterprise-ready. One subscription, every tool you need.
Built, hosted, and secured by Vinkius. You just connect and go.