GuidesTechnology

How to use Claude Code with Fabapp AI

Figures from September 17, 2026

Connecting Claude Code to your account

One command:

claude mcp add fabapp -- npx -y @fabappai/cli mcp

That is it. In the next conversation Claude Code sees Fabapp's tools and starts reading and writing in your project without you pasting anything into the chat.

It is worth understanding what it gains, because this is not just one more command. It gains your project's schema: the entities, the fields and the access rules. Without it, a wrong model id answers 404 and an invented field answers 422, and the assistant is guessing. With it, it reads before it writes.

What you need

Node 20 or newer, a Fabapp account on the Builder plan or above, and a single login:

npx @fabappai/cli login

The credential goes into the system keychain. It is never displayed, and it is revoked from one place, in the panel.

The loop

npx @fabappai/cli create "My store"
npx @fabappai/cli dev

dev assembles the app on your machine on the same template the platform publishes with, installs the dependencies and starts the local server. With it running, ask for what you would do by hand: read the definition, create the entities, write the screens. Then:

npx @fabappai/cli deploy

It uploads what you changed and publishes, in that order. Publishing before uploading would publish what is on the server, and your work would be left behind.

What is different about Claude Code

Claude Code already lives in the terminal where the app runs, so the whole loop happens in one place. Start dev, leave it running, and talk: "read the project definition and add an order model with status, customer and total". It calls the tools, writes the schema, and you watch the result in the browser without moving.

Two things trip people up. Claude Code starts the server through npx, so it needs Node on the PATH of wherever Claude Code was launched from; if you use a Node version manager, launching from a shell that has not loaded it is the most common reason the server does not come up. And a read-only credential does not show the write tools at all, on purpose. If the assistant says it cannot write, check which permission that machine was granted with npx @fabappai/cli status.

What to ask for

"Read the platform documentation before writing the schema." The field types are a closed list and anything else is refused; the access rules are the only thing between an app's data and the public.

"Read the current definition before changing it." Without the schema at hand, a wrong model id answers 404 and an invented field answers 422.

"Schema first, screens after." The SDK is generated against the data model.

Frequently asked questions

Do I pay for Claude Code and Fabapp separately?
Yes, and they are different things. The Fabapp subscription unlocks the command line from the Builder plan up. The model cost is your assistant's: when Claude Code writes the code, you pay the model, not the platform. Generating in Fabapp's Studio is the other path, and that one spends build credits.
Can Claude Code publish my app on its own?
It can, if that machine's credential has write permission. If you authorised read-only, the write tools are not even offered to it.
Does it see my integration keys?
No. The integration files say what the project has connected, so an assistant knows it exists and which operations to call, and never the credential. Connecting and rotating happens on the platform.
Start building for free
Read next
Command line and MCP in app builders: what changes when the project leaves the browserBest AI app builders in 2026