Skip to content

About

AntelopeJS DMS with the Builder and the AI assistant, and no sample data. Start your own project here.

Topics

Resources

Code of conduct

Contributing

Security policy

Stars

0 stars

Watchers

0 watching

Forks

AntelopeJS DMS template

The starting point for your own project with AntelopeJS DMS. It runs the DMS with the Builder and the AI assistant, and nothing else: no sample data, no demo application, no seeded database.

Run with Docker

You need git, Node.js 24, and Docker with Compose v2.

npx @antelopejs/core project init my-project --template dms-starter
cd my-project
docker compose up

When the log prints Local: http://localhost:3001/, open the onboarding address from the dms log and create the administrator account. See First run. The DMS keeps its own data in the MongoDB that compose.yaml starts.

Run without Docker

You need git, Node.js 24, pnpm 10.2.0, and a MongoDB server. The DMS keeps its own data in that MongoDB.

  • Node.js 24 is the version in .nvmrc.
  • pnpm 10.2.0 is the version that packageManager in package.json pins. Node.js 24 includes corepack, and corepack enable makes the pnpm command use that version.

Create the project. The command installs its packages:

corepack enable
npx @antelopejs/core project init my-project --template dms-starter
cd my-project

If you have no MongoDB on localhost:27017, start one. This is the only container you need:

docker run -d --name dms-mongo -p 127.0.0.1:27017:27017 -v dms-mongo:/data/db mongo:8

If your MongoDB listens elsewhere, set MONGODB_URL when you start the backend, for example MONGODB_URL=mongodb://localhost:27018 pnpm dev.

Run the backend and the dashboard in two terminals. In the first terminal, start the backend:

pnpm dev

Wait until the log prints Constructed modules. In the second terminal, start the dashboard:

pnpm frontend:dev

When the log prints Local: http://localhost:3001/, open the onboarding address from the pnpm dev terminal and create the administrator account. See First run. Press Ctrl C in each terminal to stop. Your data stays in MongoDB.

First run

There are no default credentials. When no administrator exists, the backend prints the address of the setup wizard at start. With Docker, it is in the dms log. Without Docker, it is in the pnpm dev terminal:

[DMS] No owner exists yet. To create it, open http://localhost:3001/onboarding?token=...

Open that address and create the administrator account. The wizard reads the setup token from the address. If you set DMS_SETUP_TOKEN, the log shows <DMS_SETUP_TOKEN> in place of the token, and you use your own value.

AI assistant

The assistant in the dashboard runs a coding agent in this project's folder, Claude Code or Codex, on your own subscription. Open the chat with the sparkle icon in the header, or press Ctrl Shift K. Its settings are under Modules > Assistant.

Without Docker. The assistant runs as you, on your machine, so it uses the logins you already have. You set no token and no API key.

  • Claude Code. If you are signed in to Claude Code on this machine, the assistant uses that login. If you are not, run claude once and sign in. The assistant runs the copy of Claude Code that @antelopejs/dms-ai installs, so it needs no claude on your PATH.
  • Codex. This project has @openai/codex 0.154.0 as a dependency, so you install nothing more. The assistant does not read a codex login from your terminal. Open Assistant > Settings and press Sign in with ChatGPT once. The sign-in is stored in ~/.config/dms-ai and serves every DMS project on your machine.

With Docker. The assistant runs inside the container, which has none of your logins. Give the container one of these before docker compose up:

  • Claude Code with your Claude subscription. On your machine, run claude setup-token and put the token in a .env file next to compose.yaml:

    CLAUDE_CODE_OAUTH_TOKEN=sk-ant-oat01-...
  • Codex with your ChatGPT plan. Open Assistant > Settings and press Sign in with ChatGPT. The page that OpenAI sends you back to does not load, because the assistant is in the container. Copy its address from the address bar and press Paste the return address.

  • An Anthropic API key, if you prefer one. Put ANTHROPIC_API_KEY=sk-ant-... in the same .env file.

Connect your own database

On Home, open the Builder with the hammer icon at the top right, choose Database, and press Connect a database. Paste a PostgreSQL, MongoDB, or RethinkDB connection string. The Builder generates a TypeScript page for each table you map, in this repository.

Without Docker, use localhost as the host of a database on your machine. With Docker, the DMS runs in a container. Use host.docker.internal as the host to reach a database on your machine. To reach a database in another container, put it on the same Compose network and use its service name.

To try the Connect flow on a sample database first, use AntelopeJS/template-dms-connect-demo.

Build pages

In the Builder, choose Page to build a page with clicks. The Builder writes what you build as TypeScript in this repository, which you can edit and commit.

Documentation

Licence

Apache-2.0

About

AntelopeJS DMS with the Builder and the AI assistant, and no sample data. Start your own project here.

Topics

Resources

Code of conduct

Contributing

Security policy

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages