# Cookwith MCP

> Cookwith MCP is an MCP server for meal planning. MCP server that generates and adapts recipes to diet, allergy and calorie goals. It is available on API, CLI and Web.

- **Kind:** MCP server
- **Platforms:** API, CLI, Web
- **Repository:** https://github.com/blaideinc/cookwith-mcp
- **MCP Registry name:** io.github.jkakar/cookwith-mcp
- **Categories:** [meal planning](https://ai-powered.com/categories/meal-planning)
- **Listing updated:** 2026-11-18

## About

The README says this MCP server gives AI clients two recipe tools: one that writes a new recipe from a plain-language request, and one that rewrites an existing recipe you supply. Constraints such as dietary restrictions, allergies, dislikes, calories per serving, protein targets and serving count can be passed to either tool.

The server is built with the Model Context Protocol, so it plugs into assistants that support MCP, and the README gives setup steps for Claude Desktop, npm installation and direct use from a client program. It can also be reached over HTTP at a cookwith.co endpoint, according to the README.

It suits home cooks and developers who want an assistant to plan meals or adjust a recipe for a specific eating pattern, such as vegetarian, gluten-free or a calorie target.

## Features

- **Recipe generation:** Creates a new recipe from a natural language description of the dish wanted.
- **Recipe transformation:** Modifies an existing recipe according to written instructions, such as a dietary change.
- **Dietary and allergy filters:** Accepts dietary restrictions, allergies and disliked foods as inputs to generation.
- **Calorie and protein targets:** Takes target calories and protein grams per serving to shape the recipe.
- **Serving adjustment:** Sets the number of servings, from 1 to 20 for generated recipes.
- **HTTP endpoint access:** The README lists an HTTP endpoint in addition to the local server.

## Good for

- Plan a quick breakfast or dinner that fits a set calorie budget
- Convert a saved recipe to gluten-free or dairy-free
- Scale a recipe up or down for a different number of guests
- Build a meal suggestion feature inside an AI assistant

## Tools

- `generate_recipe`: Generate a new recipe based on natural language instructions.
- `transform_recipe`: Transform or modify an existing recipe based on instructions.

## FAQ

### Which tools does the server expose?

The README documents two tools, generate_recipe and transform_recipe.

### How do I install it?

The README lists installation through npx, a global npm install of @cookwith/mcp-server, and a Claude Desktop configuration entry.

### What license is it under?

The README states the license is MIT.

### Can I set a limit on the number of servings?

The README says the servings parameter for generate_recipe accepts values from 1 to 20, with a default of 4.

### What are the best alternatives to Cookwith MCP?

The closest alternatives to Cookwith MCP are You2Chef, Calori, Edamam Food MCP, Ripe Tonight and Cuoco.

## Where this comes from

- Description: written from the project's README and its MCP Registry entry.

## Alternatives

- [You2Chef](https://ai-powered.com/mcp/you2chef): You2Chef is an MCP server for meal planning.
- [Calori](https://ai-powered.com/mcp/calori): Calori is an MCP server for meal planning.
- [Edamam Food MCP](https://ai-powered.com/mcp/edamam-food-mcp): Edamam Food MCP is an MCP server for meal planning.
- [Ripe Tonight](https://ai-powered.com/mcp/ripe-tonight): Ripe Tonight is a freemium MCP server for meal planning.
- [Cuoco](https://ai-powered.com/mcp/cuoco): Cuoco is an MCP server for meal planning.

---
Source: https://ai-powered.com/mcp/cookwith-mcp
