Your agent/Claude

Claude made you something.
Ask it for the link.

An artifact lives inside the conversation. A link does not. Add one connector and Claude can publish the thing it just built to a real address you can send to anyone.

The gap between “Claude made it” and “here, look”.

Claude is very good at making the thing. The trouble starts after: an artifact is stuck in your chat, a downloaded file is stuck on your laptop, and the person you wanted to show is somewhere else entirely. So the good version of it never leaves the window it was made in.

With agenthost connected, “put this online” is a whole instruction. Claude sends the files, we work out what they are and run them, and you get back an https link that works on someone else’s phone. There is no dashboard to visit afterwards and no settings page you were supposed to know about.

Updating works the same way. Ask for a change and the new version replaces the old one at the same address, usually before you have switched tabs. If you would rather not have the whole internet find it, you can name the email addresses allowed in, or ask for a password on the door.

You ask for it in a sentence.

There are no commands to learn on our side. These are whole sentences, typed the way you would say them to a person, and each one is a thing agenthost does.

put that on a real website

Publishes what Claude just made and answers with a link that works anywhere, on any phone, with https already on it.

make the heading bigger and update it

Changes the file and replaces the live version at the same address. No second link.

only let my team see it

Locks it to the email addresses you name, or puts a password on it. Search engines never see it at all.

use my own domain

Tells you the one record to paste at whoever you bought the domain from, then sorts the certificate out itself.

what have I got online right now?

Lists your projects and their addresses, and can take any of them down again.

There is more of this in the how-to guides, which are written for your agent to read as much as for you.

Whatever it built, it can publish.

You do not have to know what it is made of. Your agent works that out and tells us.

  • framework preset
  • build command
  • output directory
  • install command
  • API keys
  • credit card
  • region

Nothing here can stop and ask you a question.

Every one of these is a setting some other host would demand before it would take your files. None of them exists here, which is why your agent can go from a folder to a working link without checking in.

  • Dockerfiles
  • port settings
  • TLS certificates
  • DNS records
  • CI pipelines
  • health checks
  • scaling rules

The one thing agenthost does expect is an entrypoint per runtime, and the deploy reference is where Claude reads it before the first deploy.

How it connects.

The button opens Claude’s add-connector dialog with agenthost’s name and address already filled in. You review it and approve.

Add agenthost to Claude ↗

By hand: Settings → Connectors → Add custom connector, and paste https://agenthost.eu/mcp.

1

Add the connector

Claude opens a dialog showing the name and the address it is about to add. Nothing happens until you approve it.

2

Sign in with your email

Claude sends you to agenthost, you type the code we email you, and you approve the connection. That same step makes your account if you are new.

3

Ask for a link

Say put this online. Claude sends the files and answers with the address it is now live at.

Custom connectors work in the Claude apps, on claude.ai and in Claude Desktop. On a Team or Enterprise plan an owner adds it once under organization settings, and everyone else just signs in. The get started page walks through all of it, with the buttons and the sign-in step.

The usual ones.

Does this work on the free Claude plan?

Custom connectors are available on Free, Pro and Max, though free accounts are limited to one custom connector at a time. If you already have one connected, you would have to swap it out.

Can Claude publish an artifact it made in the chat?

Yes. An artifact is a file, and agenthost takes files. Ask Claude to put the artifact online and it sends the same content to us and hands you back the address.

Is my code sent anywhere unusual?

Claude connects to agenthost from Anthropic's servers rather than from your laptop, which is how remote connectors work for every one of them. What lands here is your files, and they are stored and served in the EU.

Our privacy policy →

What is the difference between this and Claude Cowork?

The connector is the same, and adding it once covers both. Cowork is the surface where Claude works on files for a long stretch on its own, which is where publishing the result at the end tends to matter most.

agenthost for Claude Cowork →

Is there an API key to paste anywhere?

No. agenthost signs you in through your browser, and your agent never sees a password or a key. There is nothing to copy out of one window and into another, and nothing to rotate later.

What does it cost to try?

Nothing, and there is no card. Your first ten websites are free and stay free. You only pay when the thing you built has to run code on our side, or when you want it on your own domain.

What it costs →

Can I connect more than one agent to the same account?

Yes, as many as you like. They all see the same projects, so you can build something in one and update it from another. Connecting the second one is a single click, because your browser is already signed in.

Where is this actually hosted?

In the EU, by a European company. Your files, your database and your logs stay here, and so does the billing.

Not what you use?

agenthost is a standard remote MCP server, so anything that speaks MCP can drive it. These are the ones with a page of their own.

Point Claude at it.

One connector, no card, and the thing Claude just made gets an address of its own.