The server itself exposes 45 tools, and some of them can write to your Canva account. Behind TrustGate an agent only sees the tools you allow, requests and responses are inspected, and every call is logged. Anything outside the allowlist is denied before it reaches Canva.

How is this different from connecting Canva MCP directly?

Connected directly, an agent sees every tool the server exposes and holds the credential itself. Through TrustGate you choose which tools each agent gets, and every call is logged against the agent that made it.

What if the Canva MCP API changes?

We track the Canva MCP schema, so the tool catalogue stays current. Tools you have already allowed keep working, and anything new stays denied until you enable it.

Can I give different agents different Canva tools?

Yes. Access is set per agent: one can get a handful of read tools, another the full set. You grant the whole MCP or single tools within it.

Where can I see what my agents did in Canva?

Every call is logged against the agent that made it, with the tool it used. Activity can be sent on to your own monitoring.

Do I need to run the Canva MCP server myself?

No. TrustGate connects to the Canva MCP server and sits in front of it. Your agent only talks to your TrustGate URL, never to Canva directly.

Connect Canva MCP to your AI agents, securely.

One secured endpoint for every Canva call your AI agents make, scoped per user, inspected, and logged.

Get started for free

Get a demo

Try Canva MCP now

What your agents can ask once Canva MCP is connected.

Select the tools your AI agents can call

45 tools from the Canva MCP schema.

Showing 45 tools

upload-asset-from-urlwriteUploads an image, video or audio asset from a URL into the media library.
get-assetsreadGets the details of one or more assets in the media library.
create-upload-urlwriteCreates an upload URL for adding a file to the media library.
autofill-designwriteFills a brand template with data to generate a new design.
get-brand-template-datasetreadGets the dataset that defines a brand template’s autofill fields.
get-design-datasetreadGets the autofill dataset of a design.
search-brand-templatesreadSearches the brand templates available to the user.
list-brand-kitsreadLists the brand kits available to the user.
create-design-from-brand-templatewriteCreates a new design from a brand template.
create-brand-template-draftwriteCreates a draft brand template.
publish-brand-templatewritePublishes a brand template so it can be used by the team.
comment-on-designwriteAdds a comment to a Canva design.
reply-to-commentwriteReplies to a comment on a design.
list-commentsreadLists the comments on a design.
list-repliesreadLists the replies to a comment.
export-designwriteExports a design to a file format such as PDF or an image and returns a download link.
search-designsreadSearches Canva designs.
get-designreadGets the metadata of a specific design.
get-design-pagesreadLists the pages of a design.
get-design-contentreadGets the content of a design.
get-presenter-notesreadGets the presenter notes of a presentation design.
get-export-formatsreadLists the export formats available for a design.
generate-designwriteGenerates design candidates from a prompt.
create-design-from-candidatewriteCreates an editable design from a generated design candidate.
create-designwriteCreates a new design.
get-create-design-async-jobreadGets the status and result of an asynchronous design creation job.
copy-designwriteCreates a copy of a design.
merge-designswriteMerges designs into a single design.
import-design-from-urlwriteImports a file from a URL as an editable Canva design.
resize-designwriteCreates a resized copy of a design.
get-design-thumbnailreadGets the thumbnail of a design.
helpreadReturns guidance on using the Canva MCP tools.
generate-imagewriteGenerates an image from a prompt.
get-generate-image-jobreadGets the status and result of an image generation job.
remove-backgroundwriteRemoves the background from an image.
separate-image-layerswriteSeparates an image into layers.
get-separate-image-layers-jobreadGets the status and result of an image layer separation job.
create-folderwriteCreates a folder to organise designs and assets.
list-folder-itemsreadLists the contents of a folder.
search-foldersreadSearches folders.
move-item-to-folderwriteMoves a design, asset or folder into another folder.
start-editing-transactionwriteStarts an editing transaction on a design.
perform-editing-operationswritePerforms editing operations on a design within an editing transaction.
commit-editing-transactionwriteCommits an editing transaction, saving its changes to the design.
cancel-editing-transactionwriteCancels an editing transaction, discarding its changes.

What Canva MCP exposes

Canva MCP groups its 45 tools into 7 categories, each with a suggested starting policy.

All 45 tools as JSON, ready to paste.

Category	Covers	Access	Suggested policy
Designs	Design search, details, pages, content, generation, creation, copies, merges, resizes, thumbnails and import from URL	Read / Write	Audit design creation; designs are per-user, not org-wide
Assets	Uploading and retrieving reusable assets in the media library	Read / Write	Review assets before uploading; they become reusable across designs
Folders	Folder creation, listing items and moving designs or folders	Read / Write	Audit moves; they change what collaborators can find
Brand Templates	Template search, brand kits, datasets, autofill, designs from templates and template drafts and publishing. Requires Canva Pro or above	Read / Write	Approve autofill; it can generate designs at scale
Export & Comments	Exporting designs as PDF, image or video, and comment threads	Read / Write	Review comments before posting
Images	Image generation, background removal and image layer separation	Read / Write	Review generated images before using them in published designs
Design Editing	Editing transactions: start, editing operations, commit and cancel	Read / Write	Require approval before committing edits to shared designs

Connect Canva MCP to your AI agent

Four steps, under ten minutes.

01
Sign in to NeuralTrust
Open the NeuralTrust platform, sign in and go to Getting started, then Connect your tools.
02
Choose your agent
When asked which agent you use, pick Claude Code, Codex, Cursor, Windsurf, Copilot or your own code.
03
Connect Canva
Find Canva in the tool list and connect it.
04
Connect your agent
Follow the steps on screen to add the config to your agent, then send the test prompt.

Permissions down to the single tool

You do not allow Canva MCP or block it. You allow the exact tools each agent can call, one by one.

Canva MCP	45 TOOLS AVAILABLE
Code review agent	4 tools, read only
Issue triage bot	9 tools
Platform team assistant	all 45 tools

Tool by toolConnecting the server is not the unit of access. Each tool is granted on its own.
Default denyA tool you have not enabled is never exposed to the agent.
Approval on writesTools that change or delete things can require a human to confirm.

Is Canva MCP secure?

One endpoint instead of raw credentials. Every call is checked against policy and logged.

AI agentsConnect through your MCP client configuration.	NeuralTrustApply controls based on your configuration.	Canva MCP45 documented tools. Some tools are only available on Canva Pro plans and above.
Coding assistants	Connection & credentialsper user	Designs16 tools
Internal copilots	Tool access policiesdeny by default	Brand Templates8 tools
Workflow automations	Request & response inspectioninline	Export & Comments5 tools
Compatible MCP clients	Activity monitoringalerts and logs	All other tools16 tools