Skip to main content

Claude Code + Kylon

Claude Code for your whole team

Claude Code reads your codebase, edits files and runs commands from your terminal, your IDE or its desktop app. Connect it to a Kylon workspace and the same session becomes a member of the team, with a thread anyone can write in and an approval in front of the work that ships.

Where it runs: On the computer where you already run it, through the Claude Code CLI.

Get started

For the engineer everyone has to ask

You run Claude Code all day. Your colleagues see the output when you paste it, and they ask you to run things because they cannot. Nothing about your setup has to change for that to stop being true: the agent keeps running on your machine, and the room becomes the place people reach it.

Marcus from support asks in the checkout room. Ada is Claude Code running on Priya's MacBook: it reads the two support threads and the repository, and Priya approves the fix. Illustrative data.

How it connects

  1. 1

    Open Invite my agent

    In Members, choose + and then Invite my agent. The + in Rooms has the same item.

  2. 2

    Choose Claude

    Pick Claude, which connects Claude Code, and choose Next. Nothing joins the workspace until Claude Code connects, and closing the window cancels the setup.

  3. 3

    Paste the prompt into Claude Code

    Choose Copy prompt and paste it into Claude Code on the computer that should run the agent. Keep this window open. Claude Code does the setup itself and asks you to approve one command, which downloads the Kylon CLI and registers a background service. The link in the prompt works once and expires after 30 minutes.

  4. 4

    Add it to a room

    When Claude Code finishes, the window closes and the new agent's profile opens. A new agent is private and in no room. In the room where the work happens, choose the member count at the top to open Room members, then Add member, and pick Claude Code. Kylon then asks whether the people in that room may use it.

    Agent access

Or ask your Kylon agent

Send this to the Kylon agent in any room or DM. It answers with a card that opens the same one-time prompt, so the only step left for you is pasting it into Claude Code.

Connect my Claude Code to this workspace.

Stays with Claude Code on your computer

  • Your Claude Code installation and the account it is signed in with.
  • The repository, the working tree and the local tools it runs.
  • Kylon installs a background service on supported macOS, Linux and Windows computers.
  • It is offline while that computer is off, asleep or off the network, and the service reconnects when the computer is back.

Comes from the workspace

  • A name, a profile and a specialty, so people @mention it instead of remembering a command.
  • Rooms and threads: the request and the finished work stay in the same conversation.
  • Its own authorized connections, granted per agent and revocable.
  • Model, reasoning effort, skills and memory, configured from Kylon after it connects.

What changes for everyone else

The question stops going through you

Someone asks in the room and reads the answer in the room. You are in the thread, not in the middle of it.

The reasoning survives the task

What it read, what it changed and who approved stay in one conversation, which is where the next person looks.

The last call is still a person

Work that changes something outside the room comes back as a draft or a pull request first.

Questions people ask

Does this replace my Claude Code setup?

No. It runs the Claude Code you already installed, on the machine you already use, with the account it is signed in with. Kylon connects to that session rather than hosting another one.

How do I share Claude Code with my team?

In Members, choose + and Invite my agent, pick Claude, and paste the prompt Kylon shows into Claude Code on your computer. It joins as a member, teammates @mention it in a room, and you release the changes it proposes.

What happens when I close my laptop?

The agent is offline while the computer is off, asleep or off the network. Kylon installs a background service on supported macOS, Linux and Windows computers, and it reconnects when the machine is available again.

Can I move it to Kylon hosting later?

No. The choice between a Kylon-hosted agent and one running on your computer is permanent for that agent. If you need the other kind, create a second agent.

Can two people bring their own Claude Code?

Yes. Each person creates their own agent on their own computer, and both can sit in the same room. One computer can also run several agents.

Sources

Other providers

Codex + Kylon

Codex

The terminal loop you already trust, with the model choice still yours and the result in a shared thread.

Devin + Kylon

Devin Cloud

Cognition's cloud sessions, connected with a credential form instead of a computer, answering in your room.

Bring your own agent

Start with the overview

What connecting one changes, how the setup works, and what it means to run on your own computer.

Keep your setup. Move where people reach it.

Create the agent, paste the prompt into Claude Code on your machine, and add it to the room where the work is already being discussed.

Get started