> For the complete documentation index, see [llms.txt](https://studio-docs.sandbox.game/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://studio-docs.sandbox.game/getting-started/first-time-sandbox-studio-checklist.md).

# First Time Sandbox Studio Checklist

Use this checklist to install Sandbox Studio, create a project, connect your AI agent, and complete your first edit-and-playtest loop. Open the linked guides whenever you need more detail.

**How to use this checklist:** Work from top to bottom and tick each action as you finish it. The short notes explain what to expect and how to know when you are ready to continue.

{% hint style="info" %}
**First milestone:** By the end, your project will open successfully, Sandbox Studio MCP will be connected, and you will have tested one manual edit and one AI-requested change.
{% endhint %}

{% stepper %}
{% step %}

## Understand what Sandbox Studio is

**☐** Read the overview to understand the editor, its built-in systems, and the AI-assisted workflow.

**What to expect:** This is a quick orientation. You only need a basic idea of projects, scenes, editor tools, and how the AI helps you build.

**Done when:** You understand what Sandbox Studio is and why AI changes still need to be reviewed and tested.

**Deep dive:** [What is Sandbox Studio](/getting-started/readme.md)
{% endstep %}

{% step %}

## Confirm your machine can create

**☐** Check the Creating requirements before installing Sandbox Studio; they differ from the requirements for playing a published game.

**What to expect:** Compare your operating system, memory, graphics hardware, storage, and supporting software with the creator recommendations.

**Done when:** Your computer meets the Creating requirements, or you know what must be upgraded or installed.

**Deep dive:** [System Requirements](/getting-started/system-requirements.md)
{% endstep %}

{% step %}

## Install Sandbox Studio and complete onboarding

**☐** Install Sandbox Studio, sign in with Sandbox, choose your AI assistant, and complete the environment check.

**What to expect:** The first launch may include an update, sign-in, permission prompts, assistant selection, and an environment check. Let each stage finish.

**Done when:** Sandbox Studio reaches its start screen with your account and assistant ready to use.

**Deep dive:** [Installing Sandbox Studio](/getting-started/installing-sandbox-studio.md)
{% endstep %}

{% step %}

## Create your first project

**☐** Create a project from a starter template, give it a clear name, choose where to save it, and open it.

**What to expect:** Studio may take a moment to prepare the project and load the starting scene. The template gives you a working foundation.

**Done when:** The starting scene is visible in the editor and you know where its project folder is saved.
{% endstep %}

{% step %}

## Install Cursor and sign in

**☐** Install the Cursor desktop app and sign in so its AI features are available.

**What to expect:** Choose the installer for your operating system. Sign-in may open in a browser before returning you to Cursor.

**Done when:** Cursor opens and shows that your account is signed in.

**Deep dive:** [Download and Install Cursor](/getting-started/download-and-install-cursor.md)
{% endstep %}

{% step %}

## Open the same project folder in Cursor

**☐** Use File → Open Folder in Cursor and select the exact folder created for the Sandbox Studio project.

**What to expect:** The files will appear in Cursor's Explorer and indexing may begin. Keep the same project open in Sandbox Studio.

**Done when:** The correct project files are visible in Cursor while the project remains open in Sandbox Studio.
{% endstep %}

{% step %}

## Connect Sandbox Studio MCP

**☐** Enable Sandbox Studio MCP in Sandbox Studio, connect it in Cursor, and try one small, visible request.

**What to expect:** Sandbox Studio MCP should turn green. Test it by asking the AI to create or move a simple object that is easy to check.

**Done when:** Sandbox Studio MCP is green and the requested change appears in the open scene.

**Deep dive:** [Set up and use Sandbox Studio MCP](/working-efficiently-with-ai/set-up-and-use-sandbox-studio-mcp.md)

**Video:** [How to use MCP](https://youtu.be/xp0ra7jLBIA)
{% endstep %}

{% step %}

## Complete one successful editor loop

**☐** Make one manual edit, enter Play mode, then ask the AI for one small change and test again.

**What to expect:** Keep the first loop simple: move or scale an object, play the scene, request one AI change, then replay it.

**Done when:** You can make a change, run the project without an error, and review the AI's result.

**Deep dive:** [Editor Orientation](/the-editor/editor-orientation.md)
{% endstep %}
{% endstepper %}

## Build good habits

{% stepper %}
{% step %}

## Learn the AI working habits

**☐** Read the AI guidance before asking the agent to create larger systems or groups of changes.

**What to expect:** The guides show how to plan, make small reviewable requests, choose suitable tools, and test each result.

**Done when:** Your next request has one clear outcome and you know how you will test it.

**Deep dive:** [How to Think About AI-Assisted Development](/working-with-ai-agents/how-to-think-about-ai-assisted-development.md)

**Deep dive:** [Planning before you build](/working-with-ai-agents/planning-before-you-build.md)

**Deep dive:** [Recommended Tools](/working-with-ai-agents/recommended-tools.md)
{% endstep %}

{% step %}

## Set up Git checkpoints (Optional)

**☐** Install Git and connect it to the project if you want recoverable checkpoints before larger edits.

**What to expect:** Git saves versions of your project. For now, you only need to create a repository and save one known-good checkpoint.

**Done when:** The project is connected to Git and has a checkpoint you can return to.

**Deep dive:** [Setting up Git](/source-control/setting-up-git.md)

**Deep dive:** [Connecting Git to your project](/source-control/connecting-git-to-your-project.md)
{% endstep %}

{% step %}

## Publish a shareable link (Optional)

**☐** Publish the project when it has a small playable experience that is ready for somebody else to test.

**What to expect:** Publishing prepares and uploads the project, then provides a browser link. Larger projects may take longer.

**Done when:** The link opens in a browser, the game loads, and another person can play it.

**Deep dive:** [How to Publish](/publishing/how-to-publish.md)

**Video:** [How to Publish](https://youtu.be/GxgVWHMDv-4)
{% endstep %}
{% endstepper %}

## Your First Setup Is Complete

You now have the core workflow: open the project, make a focused change, test it, review the result, and save a checkpoint before larger edits. Return here whenever you need to confirm that Sandbox Studio, Cursor, and Sandbox Studio MCP are working together.

To walk through each step in order, start with [System Requirements](/getting-started/system-requirements.md), then [Installing Sandbox Studio](/getting-started/installing-sandbox-studio.md), then [Getting Started with Sandbox Studio](/getting-started/getting-started-with-sandbox-studio.md), then [Download and Install Cursor](/getting-started/download-and-install-cursor.md).


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://studio-docs.sandbox.game/getting-started/first-time-sandbox-studio-checklist.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.