ClawKit Logo
ClawKitReliability Toolkit
Back to Registry
Official Verified media Safety 3/5

antigravity-image

Generate images using the internal Antigravity Sandbox API (Gemini 3 Pro Image). Supports text-to-image generation via internal Google endpoints.

Why use this skill?

Generate high-quality images using Gemini 3 Pro with the OpenClaw antigravity-image skill. Seamlessly create visual assets with this internal sandbox integration.

skill-install — Terminal

Install via CLI (Recommended)

clawhub install openclaw/skills/skills/houzl3416/antigravity-image
Or

What This Skill Does

The antigravity-image skill acts as a bridge between the OpenClaw agent and the internal Google Antigravity sandbox. By leveraging the gemini-3-pro-image model, this skill allows users to generate high-quality visual assets directly from text descriptions. It operates by mimicking the authentication environment of the official VS Code plugin, ensuring seamless access to the internal sandbox endpoints. The tool manages the entire lifecycle of an image request, from token retrieval in auth-profiles.json to the final file output, while automatically handling transient service errors through built-in retry mechanisms.

Installation

To integrate this skill into your local OpenClaw environment, execute the following command in your terminal: clawhub install openclaw/skills/skills/houzl3416/antigravity-image Ensure that you have an active Antigravity session authenticated. The script relies on valid local configuration files to perform the necessary OAuth handshake required by the internal API.

Use Cases

This skill is ideal for rapid prototyping of visual content. Developers can generate custom diagrams, mock-up assets, or conceptual art without leaving the chat interface. Because it uses the powerful gemini-3-pro-image model, it is particularly effective for generating highly detailed, structured imagery that requires advanced model reasoning. It is frequently used for creating documentation illustrations and custom UI elements during the development phase.

Example Prompts

  1. "Generate an image of a futuristic cityscape with flying cars and neon lights, save it as city.png"
  2. "Create a professional logo for a cybersecurity startup featuring a shield and binary patterns, output to logo.png"
  3. "Show me a high-fidelity rendering of a minimalist workspace setup with a clean desk and a large monitor, save as workspace.png"

Tips & Limitations

  • Permissions: This tool requires write access to the filesystem to save the PNG artifacts; ensure your OpenClaw working directory has appropriate permissions.
  • Authentication: If the skill fails, verify your auth-profiles.json is updated and that your Antigravity session has not expired.
  • Error Handling: The skill includes automatic retries for 503 errors, but if the sandbox remains unreachable, it is likely due to internal network restrictions or API downtime.
  • File Output: The MEDIA: tag is essential for OpenClaw to display the image; ensure you do not modify the script's output stream as this could break the automatic attachment process.

Metadata

Author@houzl3416
Stars2387
Views2
Updated2026-03-09
View Author Profile
AI Skill Finder

Not sure this is the right skill?

Describe what you want to build — we'll match you to the best skill from 16,000+ options.

Find the right skill
Add to Configuration

Paste this into your clawhub.json to enable this plugin.

{
  "plugins": {
    "official-houzl3416-antigravity-image": {
      "enabled": true,
      "auto_update": true
    }
  }
}

Tags(AI)

#image-generation#ai-art#google-api#automation#visuals
Safety Score: 3/5

Flags: file-write, external-api, code-execution