A professional AI image generation skill leveraging the GPT Image 2 model for precision typography, complex prompt following, and diverse artistic styles.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install gpt-image2
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install gpt-image2 using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is GPT Image 2?
GPT Image 2 is a specialized creative tool within the Openclaw Skills ecosystem that enables AI agents to produce high-fidelity imagery via the ClawdChat tool gateway. It is designed to overcome the common limitations of standard image models by offering exceptional accuracy in text rendering, making it a go-to choice for posters, infographics, and typography-heavy designs.
By serving as a thin client to the OpenAI GPT Image 2 model, this skill provides developers and agents with a robust framework for generating visuals that strictly adhere to multi-element prompts and specific aesthetic styles. Whether you need isometric game scenes, cinematic photography, or consistent character rendering through image-to-image workflows, this skill integrates seamlessly into your automated creative pipelines.
GPT Image 2 Use Cases
- Creating professional marketing assets like posters, menus, and infographics with accurate typography.
- Generating consistent character designs and brand-aligned visuals using identity-preserving image-to-image tools.
- Producing stylized concept art using built-in presets for Ghibli, Pixar, LEGO, and cyberpunk aesthetics.
- Automating the creation of technical assets such as exploded diagrams and retro infographics for documentation.
- Prototyping UI elements and vector illustrations for web and mobile application development.
How GPT Image 2 Works
- The AI agent sends a generation request to the gpt_image2_submit tool with parameters for prompt, style, and size.
- The request is routed through the ClawdChat gateway, deducting 300 credits from the user's account and returning a job_id.
- Because high-quality generation takes approximately 150 seconds, the agent begins a polling sequence using the gpt_image2_result tool.
- The polling tool utilizes a long-polling mechanism where the server-side waits up to 50 seconds before responding, optimizing network efficiency.
- Once the job status reaches 'done', the skill returns the final high-resolution image URLs in a flattened JSON response for the agent to process.
GPT Image 2 Setup
This skill requires the installation of the uno-cli companion to manage authentication and gateway communication within the Openclaw Skills framework.
# Install the necessary CLI dependency
clawhub install uno-cli
# Perform a one-time OAuth login to authenticate with ClawdChat
uno login
Ensure that you have sufficient credits on your ClawdChat account, as each image generation submission costs 300 credits.
GPT Image 2 Data Schema & Taxonomy
The skill returns structured data optimized for AI agent consumption. The following table describes the primary response schema:
| Property | Description | Details |
|---|---|---|
success |
Boolean | Indicates the success of the gateway call |
data.status |
String | Current job state (pending, running, done, error) |
data.job_id |
String | Unique identifier for status tracking |
data.items |
Array | Contains the generated image objects and their URLs |
meta.credits_used |
Integer | The credit cost of the specific operation |
error / hint |
String | Debugging information provided on failure |
GPT Image 2 Advanced Features
- Precise Typography: Superior text rendering capabilities that allow for specific headlines and captions to be embedded directly into generated images.
- 20 Curated Style Presets: Instant access to specialized looks including claymation, Pop Mart figurines, botanical engraving, and Ukiyo-e styles.
- Subject Preservation: Advanced image-to-image support that maintains the likeness of faces or brands across different generations.
- Asynchronous Workflow: Optimized for long-running AI tasks with built-in server-side waiting to prevent MCP timeout issues.
- Multi-Element Prompting: High adherence to complex instructions involving spatial relationships and specific object counts.
SKILL.md
Loading
Related Openclaw Skills
Terraform Cost Estimator
A financial intelligence tool that analyzes Terraform plans to provide monthly cost predictions and cloud optimization strategies.
Java Enterprise
A high-performance Java development skill focused on Spring Boot, microservices architecture, and enterprise-grade JVM optimization.
Risk Panorama - AI Internal Control Analyst
An AI-powered risk visualization and reporting engine that synthesizes complex operational data into actionable heatmaps and daily summaries.
JavaScript Pro
A professional-grade expertise module for mastering modern ECMAScript features, asynchronous patterns, and high-performance JavaScript development.
Internal Control Compliance Monthly Report Generator
An automated solution that transforms raw monthly operational and risk data into structured, management-ready compliance reports in minutes.
Tailwind Optimizer
A comprehensive auditing tool that optimizes Tailwind CSS by detecting unused utilities, identifying extraction candidates, and ensuring design system consistency.
Featured*
