polym-autocase-image-gen

Automate ChatGPT-image-2 generation through the AutoCase playground in Chrome.

8|Updated May 13, 2026
One-click install
npx skills add https://github.com/byteplus-sa/polym --skill polym-autocase-image-gen
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: polym-autocase-image-gen
Source: https://github.com/byteplus-sa/polym/tree/main/skills/polym-autocase-image-gen
Command: npx skills add https://github.com/byteplus-sa/polym --skill polym-autocase-image-gen

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It removes the manual, error-prone process of creating GPT-quality images by automating the AutoCase browser workflow for gpt-image-2 / ChatGPT-image-2.

Core Features & Use Cases

  • End-to-end Chrome automation: Drives Claude in Chrome to open AutoCase, select ChatGPT-image-2, submit prompts, and capture the generated result.
  • Text-to-image and image-to-image: Supports optional one or more local reference images for image-to-image generation.
  • Reliable completion detection and downloads: Uses DOM polling for the completion marker and saves the final image to Downloads (then moves/renames to your requested destination).

Quick Start

Ask the agent: Use polym-autocase-image-gen to generate a neon city night illustration in high quality and save it to ~/Downloads with the filename neon-city-night.png.

Frequently Asked Questions about polym-autocase-image-gen

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate gpt-image-2 generation through browser automation?

Automating gpt-image-2 generation involves driving a controlled Chrome browser to open the AutoCase playground, select the ChatGPT-image-2 model, submit prompts, and capture the generated output. This process removes manual steps by using DOM polling for completion verification and saving the final image to your filesystem.

Can I use local reference images for image-to-image generation with ChatGPT-image-2?

Yes, you can use local reference images for image-to-image generation. The automation workflow supports uploading one or more local image paths to a hidden file input in the AutoCase UI, allowing the ChatGPT-image-2 model to process reference photos alongside your text prompts.

How does the automation detect image generation completion in the AutoCase playground?

Completion detection in the AutoCase playground works by polling the browser document text for a specific total-time completion marker. This deterministic verification ensures the generated image is fully rendered before the automation saves the final output from the viewer.

Does this browser-driven image generation workflow require manual file management?

No, manual file management is not required. The automation captures the generated image and automatically saves it to Downloads. It then moves and renames the file to your requested destination path, streamlining the output retrieval process for gpt-image-2 results.

What is the best way to generate high-quality images with gpt-image-2 without manual UI interaction?

The best way to generate high-quality images without manual UI interaction is using end-to-end Chrome automation. This approach drives the browser to select ChatGPT-image-2, submit prompts, verify completion via DOM polling, and save the final output entirely hands-free.