Prefer English?
Prefer English? Switch the website language for a better experience.
Until now, AI agents connected to Webflow worked a bit like the elves in the shoemaker's workshop. You'd kick off a task, look away, and the edits would appear, already finished, on your site. Efficient, sure. Reassuring, not so much. With agent presence, Webflow pries open that black box: you can now see exactly where an agent is working, live, on the canvas, just as you'd see a colleague's cursor editing a section.
It's the logical next step after a project Webflow kicked off earlier this year with AI attribution in the site's activity log, which already let you know what an agent had changed, when, and at whose request. But a log, however thorough, only tells the story of the past. Agent presence adds the missing piece: the present.
As soon as an AI agent connects to your site through Webflow's MCP server, its working position becomes visible on the canvas in real time. Whether you're working alongside your own agent or your teammates' agents, you gain instant visibility into progress, without having to wait for the task to finish or check a log after the fact.
The rollout started on October 21, 2026 and is expected to reach every Workspace within about ten days. If an agent is already connected to your Webflow instance via MCP, there's nothing to switch on, the feature simply shows up.
| Before agent presence | With agent presence |
|---|---|
| Changes only appear once completed | Work is visible live, section by section |
| Tracking relies solely on the activity log, after the fact | Tracking happens in real time, right on the canvas |
| Hard to tell if two agents are working in parallel | Simultaneous presence is visible, just like with human collaborators |
{{webflow}}
At Mazette.co, we've been watching AI agents gain traction on Webflow for a while now, particularly through MCP. And let's be direct about it: agent presence addresses a real blind spot, not a gimmick.
Delegating tasks to an agent is one thing. Keeping trust in what it's doing, at scale, is another. On a project where several agents are working in parallel, without constant human supervision, the risk isn't that the AI makes a mistake, it's that nobody finds out until it's already live. Seeing in real time where an agent is working brings back the same collaborative instinct as watching a designer move a section live in a shared file. Nothing new in principle, but applied to AI, it's exactly what was missing to move from experimental use to production-grade use.
For an agency like Mazette.co, which often orchestrates several contributors, humans and now agents, on the same client site, this transparency changes the game on the governance side. We can finally tell a client "yes, we can see exactly what the automation is doing, live," rather than relying on a log checked after the fact. It's also a strong signal from Webflow about its vision for working with AI: not a silent replacement, but visible, traceable collaboration, and ultimately more reassuring for teams who need to be accountable.
The only minor catch is that this feature assumes you've already integrated agents via MCP into your daily workflow. If that's not yet the case, now might be the right time to start, since agent presence makes the whole experience noticeably more comfortable from the very first try.
{{seo-aeo}}
If you're already using Webflow's MCP, there's nothing to configure: agent presence activates automatically as the rollout progresses. If you're not there yet, this is a good moment to look into it, especially since real-time visibility removes much of the hesitation tied to the perceived loss of control that comes with automation.
At Mazette.co, we help our clients adopt these new workflows, from choosing the right agents to organizing work between human teams and automation, as well as a full audit of your Webflow project method. If you'd like to explore how these tools can fit into your site without losing control, get in touch, we'd love to talk it through.
Code
Over 180 companies have already stepped out of the shadows with us.
As soon as an AI agent connects to your site through Webflow's MCP server, its working position shows up live on the canvas, just like a human collaborator's cursor. You can see in real time which section, element, or page the agent is editing, without waiting for the change to be finished.
The activity log, introduced earlier by Webflow, tracks what an agent has already done, when, and at whose request. Agent presence goes further: it shows work as it happens, live on the canvas, rather than only after the fact. The two features complement each other, one for history, the other for real-time tracking.
No, Webflow is rolling out agent presence across all Workspaces at no extra cost. The only requirement is to already have an AI agent connected to Webflow through the MCP server to make use of it.
Impactful titles, polished meta descriptions, sitemaps, Schema.org, Hn structure, and audits: these are the SEO foundations that make your content readable by both Google AND generative AI. We cover both simultaneously.
With over 180 clients, our agency is one of the few in France to hold the Webflow Premium Partner label, supporting you in every area, including custom web design, sophisticated automation, conversion optimization, and SEO and GEO visibility.
We create bespoke web designs focused on performance, user journeys, and conversion. Join over 180 clients who have trusted us to elevate their brand.
