Getting Started with Google Antigravity

Learn how to access Google Antigravity, set up your workspace, navigate the interface, and create your first automated workflow.

Accessing Antigravity

Google Antigravity is available through multiple access points. Here's how to get started:

1. Sign in with Your Google Account

   Visit the Antigravity platform and sign in using your existing Google account. You'll need a Google Workspace or personal Gmail account.

2. Request Access (If Required)

   As an experimental platform, Antigravity may require you to join a waitlist or request early access. Check the platform's landing page for current availability.

3. Complete Onboarding

   Once approved, you'll be guided through a brief onboarding flow that introduces the key concepts and interface elements.

Setting Up Your Workspace

Your workspace is your central hub for managing workflows, agents, and outputs. Here's how to configure it:

Workspace Components

• Dashboard: Overview of your recent workflows, active tasks, and usage statistics.
• Workflow Editor: Visual canvas where you design and configure automated workflows.
• Agent Library: Browse and configure the AI agents available for your workflows.
• Template Gallery: Pre-built workflow templates you can customize and use immediately.
• Output Library: All artifacts generated by your workflows, organized by project.

# Recommended workspace settings

Default Model:     Gemini Pro (balanced speed/quality)
Auto-Save:        Enabled
Version History:  Keep last 30 days
Notifications:    Email + In-App
Output Format:    Google Drive (auto-sync)

Understanding the Interface

The Antigravity interface is built around a visual workflow editor. Key interface elements include:

Creating Your First Automated Workflow

Let's build a simple workflow that takes a topic, researches it, and generates a summary document.

1. Create a New Workflow

   Click "New Workflow" from the dashboard. Name it "My First Research Summary" and select the blank template.

2. Add an Input Node

   Drag an Input node onto the canvas. Set the variable name to topic and type to text. This is where you'll enter your research topic.

3. Add a Research Agent

   Drag a Research Agent node and connect it to the input. Configure it to search for recent information about the given topic.

4. Add a Summarization Node

   Drag a Text Generation node and connect it to the research agent. Set the prompt to: "Summarize the following research findings into a clear, structured document."

5. Add an Output Node

   Drag an Output node and connect it to the summarization step. Choose "Google Doc" as the output format.

6. Run the Workflow

   Click "Run", enter a topic (e.g., "Latest advances in renewable energy"), and watch the agents execute each step.

name: "My First Research Summary"
version: "1.0"

inputs:
  topic:
    type: text
    description: "Research topic to summarize"

steps:
  - id: research
    agent: web-research
    input: "{{ inputs.topic }}"
    config:
      max_sources: 10
      recency: "last_30_days"

  - id: summarize
    agent: text-generation
    input: "{{ steps.research.output }}"
    prompt: "Summarize these findings into a clear report."

output:
  format: google-doc
  destination: "My Drive/Antigravity Outputs"

Example Projects to Try

Once you're comfortable with the basics, try these starter projects: