DEDUPED Hot AI repos, turned into use cases Get 5 a day

How to revise an existing image without losing prior edits with GPT Image Skill

Edit a picture you already have, or refine the last result, while keeping references attached.

  • Content
  • Intermediate
  • 30 minutes
  • From the README
  • Codex
  • Claude Code

Who it is forA designer or product team member who has an existing image, such as a mockup or character sheet, and needs targeted changes in the same project.

The problemEach generation call is ephemeral, so a revision built from the original source silently discards earlier edits.

The outcomeAn edited PNG saved alongside the earlier output, with the edit chain kept intact.

What you need first

  • A ChatGPT or Codex plan that permits image generation (the README says Codex image generation is not available on the Free plan)
  • Node.js 22 or newer
  • Git, for installing from GitHub
  • Codex CLI signed in with ChatGPT, or a host that provides subscription-native image_gen

Steps

  1. Place the source image inside the project, or use an explicit filesystem path or @path reference to it.
  2. Ask for the change in Codex with $gpt-image Change the sky to dusk and keep the layout, using @path/to/source.png as the edit target. (inferred, check the README)
  3. For a follow-up change to the result you just made, reuse the last returned PATH as the new edit target and reattach every reference you still need.
  4. Compare the new PNG with the previous one and confirm only the requested part changed.

Prompt for your agent

Using the GPT Image Skill in <repo path>, edit <existing image path> with the change <requested change> and keep the rest of the layout. Expect a new PNG in /generated-images and a PATH=... line. Verify that the previous output was the edit target for any follow-up request.

Caveat. A pasted image in chat is not automatically passed to the nested Codex process, so the source must be a readable file path inside the project.

Share: X LinkedIn Hacker News Email Copy link

Disclaimer

Deduped lists open-source repositories that belong to other people. We do not own, maintain, audit, or vouch for any of them, and a repo appearing here is not an endorsement or a safety check. The use cases, steps, and prompts are written by AI and may be incomplete, outdated, or just wrong. Where a use case is marked "Seen in the wild", we link to public posts where someone described something similar, but we have not verified those posts and the use case itself is still our AI's own write-up. Use all of this at your own risk. Only run code, commands, or agents on systems and accounts you are authorized to use, and read the code and every command before you run it, especially anything that touches credentials, files, or the network. Nothing here is professional, security, legal, or financial advice. All trademarks and project names belong to their respective owners, and their use here does not imply affiliation. Content is provided as is, without warranties of any kind. Found a mistake, a safety problem, or something you want removed? Email [email protected] and we will look at it promptly. Full disclaimer page