Provides tools for managing designs, brands, assets, and users in Canva.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@Canva API MCP Server list my recent designs"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
A comprehensive Model Context Protocol (MCP) server for interacting with Canva's API. This server provides tools and resources for managing designs, brands, assets, and users in Canva.
Complete coverage of Canva API endpoints
Tools for all major API operations
Resources for documentation and entity details
Mock data support for testing without API credentials
Related MCP server: Anytype MCP Server
Clone this repository
Install dependencies:
npm installCopy .env.example to .env and add your Canva API credentials:
CANVA_APP_ID=your_app_id
CANVA_API_KEY=your_api_keyStart the server:
npm run devTo test the server with the MCP Inspector:
npm run inspectThis will open a web interface where you can:
Browse and test all available tools
Access resources
View server logs
get_design - Get information about a specific design
list_designs - List designs with optional pagination
get_brand - Get information about a specific brand
list_brands - List brands with optional pagination
get_asset - Get information about a specific asset
list_assets - List assets with optional filtering and pagination
upload_image - Upload an image to Canva from a URL
get_user - Get information about a specific user
list_users - List users with optional pagination
canva://overview - General API overview
canva://getting-started - Getting started guide
canva://authentication - Authentication information
canva://designs - Designs API documentation
canva://brands - Brands API documentation
canva://assets - Assets API documentation
canva://users - Users API documentation
canva://design/{designId} - Information about a specific design
canva://brand/{brandId} - Information about a specific brand
canva://asset/{assetId} - Information about a specific asset
If no API credentials are provided, the server will use mock data for all responses. This is useful for testing and development.
MIT
Inputs: assetId (required).
Inputs: brandId (required).
Inputs: designId (required).
Inputs: userId (required).
Inputs: type, limit, startAfter.
Inputs: limit, startAfter.
Inputs: limit, startAfter.
Inputs: limit, startAfter.
Inputs: url (required), title, brandId.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 9 tools
The get/list pairs for each resource are distinct and easy to differentiate. The only potential confusion is 'upload_image' versus 'get_asset'/'list_assets', but the verb 'upload' clearly indicates a different action. Without descriptions, the boundaries between asset, design, user, and brand are clear from the names.
Most tools follow a consistent verb_noun pattern with 'get_' for singular resources and 'list_' for plural collections (e.g., get_user, list_users). The exception is 'upload_image', which uses a different verb and breaks the otherwise predictable pattern. The mix is minor but noticeable.
With 9 tools, the count is within the typical well-scoped range. Each resource has a get and list tool, which is reasonable for a read-focused server. The count feels appropriate for the apparent scope, though the lack of mutating tools may make the set feel slightly thin for a full API server.
The tool surface is severely incomplete for a Canva API. Only get and list operations are present for assets, users, designs, and brands, with no create, update, or delete functionality. 'upload_image' provides some creation capability for assets, but the rest of the resources have no way to create or modify them, leaving significant gaps for agents.
The Canva MCP server connects AI assistants (like Claude, ChatGPT, and Cursor) to Canva's API, enabling them to create and manage designs directly within chat conversations. Key capabilities include generating new designs from prompts, autofilling templates, searching and resizing existing designs, importing files from URLs, exporting designs as PDFs or images, and managing folders and comments without switching between tools.
A comprehensive Model Context Protocol (MCP) server that enables AI assistants to interact with yo…
Create, search, edit, and export designs and assets from Canva.
An MCP server that provides an API to LLMs to manage their JumpCloud resources.
