All three edit the same project source and deploy to the same platform, so you can move between them freely — start something in the app, refine it from your editor, or vice versa.
Luno Builder is not Luno, the read-only assistant everyone gets in the chat dock: Luno answers questions about what you can see and never creates or changes anything. See Chat.
Start a chat
Open Luno Builder, the home of Studio (/studio); builders, admins and owners have Studio. Each chat session is bound to one Space (project) in your active organization — to work on a different one, start a chat there. With no Space yet, your first message creates one.
Describe the outcome you want rather than choosing a primitive up front, the same way you would prompt a coding agent:
How it works
Behind the chat, Luno Builder works in a hosted session with theluno CLI and a checkout of your project’s managed Git repository — the same source a coding agent edits locally.
- It edits the shared draft. Every project repository has two branches:
mainholds the deployed, live state, andluno/draftis the shared work-in-progress draft. Luno Builder saves its edits toluno/draft(and again when its turn ends). Teammates and their sessions share the same draft. - The web app renders the deployed state. Workflow canvases, agent pages, and run history show what’s live. Draft edits appear in the Space’s Files tab in Studio (which tracks the draft in real time) until they’re deployed.
- Deploying promotes the draft. When the agent runs
luno deploy, the draft is built, health-checked, and promoted tomainand the live runtime. Until then, changes exist only on the draft. - It’s already authenticated. Sessions carry your project’s API credentials. Luno Builder never asks you to log in and never runs
luno auth login.
Connect integrations from chat
When a build needs an app connection (Slack, Google, Linear, an API key), Luno Builder shares a connect link. Opening it launches the in-app connect dialog, where you approve OAuth or enter the key and choose the scope. The agent then verifies the connection and continues. Secrets never pass through the chat — you enter them in the connect flow, and they’re stored as credentials like any other connection.Working alongside coding agents
Luno Builder and coding agents are collaborators on the same project, not alternatives you must choose between once:- A coding agent working locally picks up Luno Builder edits with
luno pull(which fetches the shared draft) and publishes its own work throughluno deploy. - Luno Builder sees local work as soon as it’s deployed or on the draft.
- Everything either one builds is ordinary TypeScript in the project — reviewable, testable, and maintainable by the other.
If you are an agent reading this in a hosted session
Reading this from inside a Luno OS-hosted session (Luno Builder, or an MCP workspace)? The local-setup docs don’t apply to you:- You are already authenticated — never run
luno auth login, and skip install/luno init/projects linkbootstrap steps. - Your edits go to the shared draft (
luno/draft): Luno Builder saves them with its save tool (never with git commands), MCP workspaces with the file tools andexec_command. Runluno typecheckbefore you finish — a broken shared draft affects everyone. - You cannot open a browser. For app connections, use the
connect_apptool (orluno connect <slug> --print-url) and share the link with the user. - Draft edits are not live until
luno deploypromotes them — say so when you finish without deploying.
xiaomi/mimo-v2.6-pro); the Luno assistant and other chats start on DeepSeek V4.1 Flash (deepseek/deepseek-v4.1-flash). When the organization turned that default off, the organization’s default model is used, else the first enabled one. The picker remembers your last model and Space. Prompts with a model the organization doesn’t offer are rejected with 403.
Self-hosting: Luno Builder runs on your platform’s model gateway and sandbox provider, so a self-hosted deployment needs a model provider key configured (for example
AI_GATEWAY_API_KEY or ANTHROPIC_API_KEY) before chat can answer.Next steps
CLI for agents
The local editor path: headless CLI, JSON output, deploy-first loop.
MCP for agents
Build from ChatGPT, Claude, or any MCP client in a hosted workspace.
Deploy a project
The draft/main model and what deploy actually does.
Connect credentials
Connect apps, API keys, and OAuth accounts.