Installation
Install the Appling CLI, open the IDE, connect an AI assistant and prepare the toolchains for code generation.
Appling is a set of npm packages. You only install what you use: the CLI is enough to write and check models; the IDE, the MCP server and the code generators come on top.
Requirements
- Node.js 24 or later.
- For code generation only: uv for the Django target, Elixir 1.17 or later with the
igniter_newarchive for the Ash target. See the Django and Ash pages for the exact requirements.
The CLI
Add the CLI to your project as a development dependency:
npm install --save-dev @appling/cli
Or run it without installing:
npx @appling/cli check models
The package installs the appling command. Check that it works:
npx appling check
Without a path it checks every .appling file below the current folder. See the CLI reference for every command.
The IDE
The IDE runs locally and opens in your browser. It needs no installation:
npx @appling/ide models
Then open http://127.0.0.1:4410. Pass --port to use another port. See the IDE.
AI assistants
The MCP server works with Claude Code, Claude Desktop and any other MCP client. For Claude Code, install it in the project and register it:
npm install --save-dev @appling/ai
claude mcp add appling -- npx appling-mcp .
The IDE chat needs an API key in the environment (ANTHROPIC_API_KEY or OPENAI_API_KEY). See AI.
Code generators
Install the extension for your target next to the CLI, then list it in appling.config.json:
npm install --save-dev @appling/codegen-django
{
"generate": [
{ "extension": "@appling/codegen-django", "out": "build/shop", "options": { "name": "shop" } }
]
}
See code generation.
Editor support
Any editor with a language server client can use appling-language-server --stdio from @appling/lang for diagnostics, completion, rename and formatting. See editors.

