# AI use cases for GPT Image Skill (GENEXIS-AI/gpt-image-skill) Repo: [GENEXIS-AI/gpt-image-skill](https://github.com/GENEXIS-AI/gpt-image-skill) | Use cases: 5 | Page: https://dedupedai.com/use-cases/genexis-ai-gpt-image-skill/ GPT Image Skill is an installable skill that lets Codex, Claude Code, or Google Antigravity generate and edit GPT images through a ChatGPT subscription instead of the Images API. This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL. - Generate images from a text prompt in Codex: https://dedupedai.com/use-cases/genexis-ai-gpt-image-skill/generate-images-from-text-prompt/ - Revise an existing image without losing prior edits: https://dedupedai.com/use-cases/genexis-ai-gpt-image-skill/revise-existing-image-keep-edits/ - Batch-create several distinct poster concepts at once: https://dedupedai.com/use-cases/genexis-ai-gpt-image-skill/batch-distinct-poster-concepts/ - Make transparent-background icons for app projects: https://dedupedai.com/use-cases/genexis-ai-gpt-image-skill/transparent-background-app-icons/ - Keep one character consistent across new scenes: https://dedupedai.com/use-cases/genexis-ai-gpt-image-skill/consistent-character-across-scenes/ ## How to generate images from a text prompt in Codex with GPT Image Skill Repo: [GENEXIS-AI/gpt-image-skill](https://github.com/GENEXIS-AI/gpt-image-skill) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Install the skill once, then create a finished GPT image from one natural-language request. **Who it is for:** A solo creator or developer with a ChatGPT subscription who needs a quick illustration for a blog post, slide, or project page and works inside Codex or Claude Code. **The problem:** Image generation usually means a separate tool, a separate bill, or an API key. The user wants images without leaving the coding agent. **The outcome:** A PNG saved in /generated-images inside the active project, returned as a PATH and an inline Markdown link. ### Steps 1. Paste the install prompt from the README into Codex, Claude Code, or Google Antigravity, and follow the Sign in with ChatGPT device authorization when asked. 2. Wait for the setup to report readiness. The README states that no image is generated during setup. 3. In Codex, type `$gpt-image Create a cozy reading room at sunset, 16:9, high quality.` In Claude Code, use `/gpt-image` with the same request. 4. Open the returned PATH in /generated-images and check the image against your request. ### Prompt to paste into your agent ```text Install GPT Image Skill from https://github.com/GENEXIS-AI/gpt-image-skill using the AGENT_INSTALL.md contract, then generate one image of at with high quality. Expect a PNG under /generated-images and a PATH=... line in the output. Verify the file exists and that the output did not use OPENAI_API_KEY or the Images API. ``` ### Caveat Generation draws on included ChatGPT or Codex usage, so a Free plan account cannot complete these steps. Risk, credentials: Setup asks you to complete Sign in with ChatGPT device authorization, so only approve it on a machine and account you control. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/genexis-ai-gpt-image-skill/generate-images-from-text-prompt/ ## How to revise an existing image without losing prior edits with GPT Image Skill Repo: [GENEXIS-AI/gpt-image-skill](https://github.com/GENEXIS-AI/gpt-image-skill) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Edit a picture you already have, or refine the last result, while keeping references attached. **Who it is for:** A 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 problem:** Each generation call is ephemeral, so a revision built from the original source silently discards earlier edits. **The outcome:** An edited PNG saved alongside the earlier output, with the edit chain kept intact. ### 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 to paste into your agent ```text Using the GPT Image Skill in , edit with the 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. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/genexis-ai-gpt-image-skill/revise-existing-image-keep-edits/ ## How to batch-create several distinct poster concepts at once with GPT Image Skill Repo: [GENEXIS-AI/gpt-image-skill](https://github.com/GENEXIS-AI/gpt-image-skill) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Delegate five different poster designs and generate them in one bounded parallel run. **Who it is for:** A marketing or event lead who needs several competing visual directions for one campaign by the end of the day. **The problem:** Generating each concept one at a time is slow, and reusing one prompt produces near-duplicates. **The outcome:** A set of distinct PNG concepts in /generated-images, each with its own job ID and filename, generated with bounded parallelism. ### Steps 1. Describe the shared subject, brand, text, and aspect ratio once, then ask for a specific number of different poster concepts. 2. Let the agent write one complete, meaningfully different image prompt per concept while keeping the shared constraints. 3. Allow the batch to run. The README says default concurrency is 2 and the maximum is 4, with one ChatGPT auth check for the batch. 4. Review each returned PATH and keep the concepts you want to refine with the revision workflow. ### Prompt to paste into your agent ```text Using GPT Image Skill in , create distinct poster concepts for at . Expect one PNG per concept in /generated-images, each with its own filename. Verify that the prompts differ in composition and not only in wording. ``` ### Caveat Batch runs still consume included ChatGPT or Codex usage and remain subject to plan and workspace limits. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/genexis-ai-gpt-image-skill/batch-distinct-poster-concepts/ ## How to make transparent-background icons for app projects with GPT Image Skill Repo: [GENEXIS-AI/gpt-image-skill](https://github.com/GENEXIS-AI/gpt-image-skill) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Generate an icon with a transparent background and check that the PNG really has alpha. **Who it is for:** A developer building a small app or website who needs logo-style icons that sit cleanly on any background. **The problem:** Many image tools return a solid background, so the icon needs manual cutting before it can be used. **The outcome:** A transparent PNG icon in /generated-images, passed through the bridge's alpha check. ### Steps 1. Ask for the icon with the transparency request stated, for example `$gpt-image Create a flat blue robot app icon with a transparent background, 1:1, high quality.` (inferred, check the README) 2. Use `--background transparent` only when the transparency is requested, as the README advises. 3. Open the returned PNG and confirm the background is transparent in an image viewer that shows alpha. 4. Copy the icon into your app's asset folder and reference it from there. ### Prompt to paste into your agent ```text Using GPT Image Skill in , generate a flat icon of at 1:1 with a transparent background. Expect a transparent PNG under /generated-images with a PATH=... line. Verify the file has an alpha channel before it is used in . ``` ### Caveat The README does not promise identical pixel dimensions, since exact sizes may vary with built-in image generation. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/genexis-ai-gpt-image-skill/transparent-background-app-icons/ ## How to keep one character consistent across new scenes with GPT Image Skill Repo: [GENEXIS-AI/gpt-image-skill](https://github.com/GENEXIS-AI/gpt-image-skill) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Reuse a character reference file to place the same person or mascot into several new scenes. **Who it is for:** A children's book author, game artist, or brand designer who needs one recurring character shown in many settings. **The problem:** Describing a character in text drifts between images, while the same reference file keeps identity more stable. **The outcome:** A set of PNG scenes featuring the same reference character, each in its own file. ### Steps 1. Save the character reference as a PNG, JPEG, or WebP file inside the project, for example at references/character.png. 2. Ask for a scene with `/gpt-image Use @references/character.png as the character reference and place it in a rainy city, 9:16, high quality.` in Claude Code. 3. Repeat the request for each scene, keeping the same reference path in every job. 4. Check that each returned PATH shows the same character before you use the set. ### Prompt to paste into your agent ```text Using GPT Image Skill in , generate scenes using as the reference in each job, with the scene descriptions . Expect one PNG per scene under /generated-images. Verify the reference file is attached in every job, not replaced by a text description. ``` ### Caveat The README says a host-attached image is not automatically inherited by the nested codex exec process, so the reference must be a readable file in the project. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/genexis-ai-gpt-image-skill/consistent-character-across-scenes/