Webflow MCP Server: What It Does and How to Connect Webflow to AI in 2026
Yes — Webflow has an official MCP server. It’s built and maintained by Webflow, open source under an MIT license, and hosted remotely at mcp.webflow.com — one of the first website platforms to ship a native MCP integration. It lets any MCP-compatible AI tool read and update your sites, CMS collections, and pages through natural language. So if you’re searching “Webflow MCP,” the connection you want exists and it’s real.
The thing worth knowing before you set it up: an MCP server hands your Webflow project to an AI inside a conversation you start. It’s a doorway, not a worker. Nothing watches your site for you, nothing fires when a form comes in or a CMS item changes, and nothing runs while the chat is closed. Here’s exactly what the Webflow MCP does, how to turn it on, where it stops — and what to use when you want Webflow work that runs on its own.
What the Webflow MCP server does
Model Context Protocol (MCP) is the open standard that lets an AI client — Claude, ChatGPT, Cursor, Codex, and others — talk to an outside app through a shared interface. Webflow’s official server (webflow/mcp-server on GitHub) exposes two categories of tools:
- Data API tools — the remote, no-canvas-needed set: list and get sites, read and write CMS collections and items, manage pages and their SEO settings, handle assets and forms, add custom code, and publish to your domains.
- Designer API tools — create and modify elements, styles, components, and variables directly on the visual canvas. These require the Webflow Designer open with the MCP Bridge App connected, so they aren’t fully headless the way the Data API tools are.
For most people, the Data API side is what “Webflow MCP” means day to day. With it connected, an AI client can:
- Read your site and CMS — “list the items in the Blog Posts collection and show which are unpublished” answered from the live project.
- Create and update content — add a CMS item, edit a page’s meta title, update a field without opening the Designer.
- Publish — push changes live to your staging or custom domain on request.
- Reason across your content — spot posts missing a hero image, summarize a collection, draft field values from existing entries.
It’s genuinely useful for ad-hoc work: ask a question about your site, get an answer grounded in the real project, make a change on the spot.
How to set up the Webflow MCP server
The remote server is the quick path — no code, no hosting:
- In your AI client’s connector settings, add a remote MCP server pointing at
https://mcp.webflow.com/sse(some clients use the/mcpendpoint — Webflow’s docs list the exact string for each). - Authorize it against your Webflow account through the OAuth prompt, selecting the sites the AI is allowed to touch — you’re granting it the same access your login has to those sites.
- Confirm the tools appear in the client, then start a chat and ask it to read or update a CMS item.
If you want the Designer tools too, open the site in the Webflow Designer and launch the published Webflow MCP Bridge App, then wait for it to connect. There’s also a local option that authenticates with a Webflow API token via environment variables, but the remote OAuth flow is the one non-developers want. Webflow’s MCP docs cover both.
Where the Webflow MCP stops
None of this is a knock on MCP — it’s just the shape of the protocol. Four limits show up the moment you want more than a conversation:
- It only works inside a chat you start. Close the window and nothing happens. The AI doesn’t watch Webflow; it waits for you to ask. And the Designer tools go dark the moment you close the canvas or the Bridge App.
- No triggers. A form submission, a new CMS item, a comment left on the site — none of these can start anything through MCP. There’s no “when this happens in Webflow, do that.”
- It’s one app at a time. The Webflow MCP knows Webflow. Getting a form lead into your CRM, a Slack channel, and an email tool means wiring up (and authing) a separate MCP server for each, then hoping your client can juggle them in one turn.
- You own the plumbing and the scopes. OAuth tokens, refresh, which sites you authorized, and the blast radius of write-and-publish access are all on you.
So the Webflow MCP is a great way to ask your site things and make one-off edits or publishes. It is not a way to make Webflow run — to have work happen on a schedule or in reaction to an event, across the other tools your site touches.
Running Webflow work that doesn’t need a chat open
That “run on its own, across apps” gap is exactly where Carly fits. Carly connects to Webflow natively — no MCP server to host, no OAuth plumbing to maintain, no Bridge App to keep open — and to the ~260 other apps it supports natively, plus anything with a public API through your own key. The difference from MCP is the important part: Carly’s workflows are triggered and scheduled, so Webflow work happens whether or not anyone has a chat window open.
A few things that MCP can’t do but a Carly workflow can:
- When a form is submitted on your Webflow site → create the contact in your CRM, post it to the
#leadsSlack channel, and send an intro email — automatically, the moment it comes in. - Every morning → find CMS items scheduled to go live today, publish them, and send the list to your editor.
- When a new row lands in a Google Sheet → create the matching Webflow CMS item and publish it, then mark the row done.
The non-AI steps — the moving, matching, and routing between apps — are free and unlimited, the Zapier-style backbone of the workflow. The AI steps (drafting, summarizing, deciding) start at $35/month. You describe the outcome in plain language and Carly wires up the Webflow connection and everything downstream.
If you just want to interrogate or edit your site from a chat, Webflow’s official MCP server is the right tool and it’s free to connect. If you want Webflow to actually do things — on a trigger, on a schedule, across every app a form lead or CMS item flows through — that’s the job MCP wasn’t built for, and it’s the one Carly was.
FAQ
Does Webflow have an official MCP server?
Yes. Webflow maintains an official, open-source (MIT-licensed) MCP server hosted remotely at mcp.webflow.com. It gives MCP-compatible AI tools read/write access to your sites, CMS, and pages via a Data API, plus canvas-editing Designer tools that require the Webflow Designer open with the MCP Bridge App connected.
Is the Webflow MCP server free? Connecting it is free; you’re authorizing an AI client against your existing Webflow account and its access to your sites. You still need whatever Webflow plan your sites and CMS live on, and API rate limits apply.
Can the Webflow MCP trigger automations? No. MCP is request/response inside an AI chat — it has no triggers and nothing runs when the conversation is closed. For event- or schedule-driven Webflow work across apps, such as reacting to a form submission, you need a workflow tool like Carly rather than an MCP server.
Can I connect Webflow to AI without coding or hosting a server? Yes. You don’t have to touch MCP or keep a Bridge App open. Carly connects to Webflow for you and lets you build the automation in plain language — describe what you want to happen and it wires up the site and the other apps involved, with no server to host and no code to write.
Ready to automate your busywork?
Carly schedules, researches, and briefs you—so you can focus on what matters.
See what people say
"Before Carly, I relied on a Calendly link, but the whole process felt impersonal and not very professional. Carly changed that by handling all the back-and-forth, so I'm no longer stuck in endless email threads trying to line up schedules.
Now Carly reaches out to candidates, shares my real-time availability, lets them pick a slot, then sends a Zoom link and drops it straight into my calendar. She sends reminders to both of us before each call, which has significantly reduced no-shows and last-minute confusion.
On top of scheduling, Carly acts like a full executive assistant, sending me my schedule the night before so I can prepare for each call. It reminds me of the old x.ai assistant, but Carly is noticeably smarter, faster, and better suited to my healthcare recruitment business."
