Cookwith Mcp

io.github.jkakarPublished by GitHub account jkakarv1.0.2Updated Nov 10, 2026

AI-powered recipe generation and transformation tools by Cookwith

VerifiedSTDIODesktop onlyLocation & LifestyleAI & ML

Overview

AI-generated overview

A local MCP server that lets an assistant generate new recipes and adapt existing ones to dietary needs, allergies and nutrition targets using Cookwith's…

What you can do
  • Generate recipes from a natural-language prompt, with optional dietary restrictions, allergies, dislikes, calorie and protein targets, and serving counts.
  • Transform an existing recipe (title, description, ingredients, instructions, servings) according to plain-language instructions such as making it vegetarian or gluten-free.
  • Adjust recipes for new calorie, protein or serving targets.
When to use it
Useful when you want an assistant to draft meal ideas or rework a recipe you already have for dietary restrictions, allergies, portion sizes or nutrition goals, without leaving the chat.
Requirements
Runs as a local process on the desktop, launched via npx from the npm package named in the README; Node.js and network access are implied. No accounts, API keys, environment variables or headers are declared.
Before you install
Recipe requests, including dietary and allergy details, are sent to Cookwith's external culinary AI service, so treat personal health information accordingly. Generated recipes are AI output and should be checked for allergens and food-safety before use.

Installation

In SourceWeft

  1. Open Cookwith Mcp in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.

Other MCP clients

Follow the launch instructions in the repository.

README

Cookwith MCP Server

An MCP (Model Context Protocol) server that provides AI-powered recipe generation and transformation tools using Cookwith's advanced culinary AI.

Features

  • Recipe Generation: Create custom recipes from natural language descriptions
  • Recipe Transformation: Modify existing recipes based on dietary needs, serving sizes, or other requirements
  • Dietary Support: Handle allergies, dietary restrictions, and nutritional goals
  • Smart Adaptations: Adjust for calories, protein targets, and serving counts

Installation

Via MCP Registry

bash
npx @modelcontextprotocol/create-server install @cookwith/mcp-server

Via npm

bash
npm install -g @cookwith/mcp-server

For Claude Desktop

Add to your Claude Desktop configuration:

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json Windows: %APPDATA%\Claude\claude_desktop_config.json

json
{  "mcpServers": {    "cookwith": {      "command": "npx",      "args": ["@cookwith/mcp-server"]    }  }}

Available Tools

generate_recipe

Generate a new recipe based on natural language instructions.

Parameters:

  • prompt (string, required): Natural language description of the desired recipe
  • dietaryRestrictions (array): Dietary restrictions (e.g., vegetarian, vegan, gluten-free)
  • allergies (array): Ingredients to avoid due to allergies
  • dislikes (array): Foods the user doesn't like
  • calories (string): Target calories per serving
  • protein (string): Target protein in grams per serving
  • servings (number): Number of servings (1-20, default: 4)

Example:

javascript
{  "prompt": "A healthy pasta dish with lots of vegetables",  "dietaryRestrictions": ["vegetarian"],  "calories": "500",  "servings": 2}

transform_recipe

Transform or modify an existing recipe based on instructions.

Parameters:

  • recipe (object, required): The recipe to transform
    • title (string): Recipe title
    • description (string): Recipe description
    • ingredients (array): List of ingredients
    • instructions (array): Cooking instructions
    • servings (number): Number of servings
    • Additional optional fields for nutrition, timing, etc.
  • instructions (string, required): How to transform the recipe
  • calories (string): New target calories per serving
  • protein (string): New target protein per serving
  • servings (number): New number of servings

Example:

javascript
{  "recipe": {    "title": "Classic Spaghetti Carbonara",    "description": "Traditional Italian pasta dish",    "ingredients": ["400g spaghetti", "200g guanciale", "4 eggs", "100g pecorino"],    "instructions": ["Cook pasta", "Fry guanciale", "Mix eggs and cheese", "Combine"],    "servings": 4  },  "instructions": "Make it vegetarian and reduce calories",  "calories": "400"}

Usage Examples

With Claude

Once configured, you can use natural language to interact with the tools:

"Generate a healthy dinner recipe for 2 people with chicken and vegetables, around 500 calories per serving"
"Transform this pasta recipe to be gluten-free and dairy-free"

Programmatic Usage

javascript
import { Client } from '@modelcontextprotocol/sdk';
const client = new Client({  name: 'my-app',  version: '1.0.0'});
await client.connect('npx', ['@cookwith/mcp-server']);
// Generate a recipeconst result = await client.callTool('generate_recipe', {  prompt: 'Quick and healthy breakfast',  calories: '350',  servings: 1});

Development

Building from Source

bash
git clone https://github.com/blaideinc/cookwith-mcpcd cookwith-mcpnpm installnpm run build

Running Locally

bash
npm start

Testing

bash
npm test

API Endpoint

The MCP server can also be accessed via HTTP at:

  • Production: https://cookwith.co/api/mcp
  • Development: http://localhost:3000/api/mcp

License

MIT

Support

About Cookwith

Cookwith is an AI-powered cooking platform that generates personalized recipes based on your preferences, dietary restrictions, and taste profile. Learn more at cookwith.co.

Source: README.md at commit 294d135

Tools

0
SourceWeft has not started this package in its sandbox yet, so its tools are not known.

Version history

1
  1. v1.0.2LatestSep 16, 2026

Score

50% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 days25 pts

    Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.

  • Auth declared10 pts

    Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.

  • Tools listed15 pts

    Answer tools/list without a credential, or ship a package that starts in a sandbox without one.

  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts