Skip to content

wuyoscar/GPT-Image2-Skill / skills/gpt-image

gpt-image

Flagged

Generate or edit images through the packaged CLI and Reference Gallery, defaulting to GPT Image 2.6 Flare. Use for posters, typography, reference edits, inpainting, native GIFs, sprite animation, image-led websites, and explicit Image 2/2.5/2.6 requests.

FileKindSize
SKILL.mdskill md11.8 KB
agents/openai.yamlconfig350 B
references/craft.mdmarkdown19.4 KB
references/gallery-anime-and-manga.mdmarkdown17.4 KB
references/gallery-architecture-and-interior.mdmarkdown9.8 KB
references/gallery-beauty-and-lifestyle.mdmarkdown2.4 KB
references/gallery-brand-systems-and-identity.mdmarkdown3.3 KB
references/gallery-character-design.mdmarkdown1.9 KB
references/gallery-cinematic-and-animation.mdmarkdown5.9 KB
references/gallery-cinematic-film-references.mdmarkdown7.4 KB
references/gallery-data-visualization.mdmarkdown6.4 KB
references/gallery-edit-endpoint-showcase.mdmarkdown3.0 KB
references/gallery-events-and-experience.mdmarkdown2.5 KB
references/gallery-fashion-editorial.mdmarkdown8.7 KB
references/gallery-fine-art-painting.mdmarkdown5.9 KB
references/gallery-gaming.mdmarkdown12.8 KB
references/gallery-illustration.mdmarkdown4.8 KB
references/gallery-infographics-and-field-guides.mdmarkdown17.0 KB
references/gallery-ink-and-chinese.mdmarkdown2.5 KB
references/gallery-isometric.mdmarkdown4.6 KB
references/gallery-more-illustration-styles.mdmarkdown6.6 KB
references/gallery-official-openai-cookbook-examples.mdmarkdown3.8 KB
references/gallery-photography.mdmarkdown8.6 KB
references/gallery-pixel-art.mdmarkdown1.8 KB
references/gallery-product-and-food.mdmarkdown5.9 KB
references/gallery-research-paper-figures.mdmarkdown28.2 KB
references/gallery-retro-and-cyberpunk.mdmarkdown4.8 KB
references/gallery-scientific-and-educational.mdmarkdown8.6 KB
references/gallery-screen-photography.mdmarkdown3.2 KB
references/gallery-tattoo-design.mdmarkdown4.3 KB
references/gallery-technical-illustration.mdmarkdown7.0 KB
references/gallery-typography-and-posters.mdmarkdown14.9 KB
references/gallery-ui-ux-mockups.mdmarkdown6.3 KB
references/gallery-watercolor.mdmarkdown2.3 KB
references/gallery.mdmarkdown4.8 KB
references/models.mdmarkdown8.9 KB
references/openai-cookbook.mdmarkdown46.1 KB
references/openai-image-2.5-editing.mdmarkdown3.5 KB
references/openai-image-2.5-generation.mdmarkdown2.9 KB
references/openai-image-2.5-layout-and-text.mdmarkdown2.9 KB
references/openai-image-2.5-migration.mdmarkdown2.8 KB
references/openai-image-2.5.mdmarkdown1.4 KB
references/templates-gpt-image-2.5.mdmarkdown6.6 KB
scripts/generate.pyexecutablescript2.5 KB

references/openai-image-2.5-editing.md

View on GitHub
# GPT Image 2.5: references and preservation

Read for reference edits, translation, inpainting and transparent cutouts.
Source check: **2026-09-09**;
[provenance and validation boundary](openai-image-2.5.md#source-boundaries).

## Official examples to consult selectively

- [Translation with layout preservation](https://developers.openai.com/api/docs/guides/image-prompting?model=gpt-image-2.5#translate-while-preserving-layout)
- [Identity and clothing](https://developers.openai.com/api/docs/guides/image-prompting?model=gpt-image-2.5#preserve-identity-and-change-clothing)
- [Combining references](https://developers.openai.com/api/docs/guides/image-prompting?model=gpt-image-2.5#combine-references)
- [Transparent product cutouts](https://developers.openai.com/api/docs/guides/image-prompting?model=gpt-image-2.5#create-a-transparent-product-cutout)
- [Character consistency across turns](https://developers.openai.com/api/docs/guides/image-prompting?model=gpt-image-2.5#keep-a-character-consistent)

Improved preservation is not a pixel lock. Repeated edits can drift; exact
unchanged regions require compositing rather than relying on prompts alone.
[Official migration caution](https://developers.openai.com/api/docs/guides/image-prompting?model=gpt-image-2.5#migrate-an-existing-workflow)

## Local workflow using the existing CLI

1. Inspect the actual inputs. Assign each a role: base scene, identity, garment,
   style or inserted object. Keep this order consistent with repeated `-i` flags.
2. Separate the requested change from invariants; do not turn a small edit into a
   restyle. Consult [craft.md](craft.md) section 13 only if this needs clarification.
3. Keep the selected model explicit. Use [models.md](models.md#parameters) for
   API settings; do not infer that 2.5 requires `--input-fidelity high`.
4. Compare the changed region and the rest of the image. For another authorized
   edit, pass the actual approved output through `-i`; the CLI does not retain
   a conversation or attach an image merely because its filename is in the prompt.

| Task | Inspect before accepting |
|---|---|
| Translation | Approved replacement copy, untranslated remnants, original layout and logos |
| Try-on / identity edit | Face, expression, pose, garment details and plausible body geometry |
| Multi-reference composition | Correct source for each inserted element, scale, perspective and lighting |
| Cutout | Product outline/labels, edge halos, hair/glass/shadows and real alpha transparency |
| Multi-step edit | Current requested change plus accumulated drift from the original invariants |

## Masks and transparent output

A mask is edit guidance, not a guaranteed hard pixel boundary. With multiple
inputs it applies to the first image. Match mask dimensions to that image and
check its alpha channel; use `-i` plus `-m`. See the
[official mask requirements](https://developers.openai.com/api/docs/guides/image-generation#edit-an-image-using-a-mask)
for input constraints.

For cutouts, use `--background transparent --format png` (or WebP), not JPEG.
Inspect the decoded alpha channel; a painted checkerboard is not transparency.
The prompt's preservation list alone does not establish that anything was
preserved. If the job requires exact untouched pixels, explain the need for a
separate compositing step; do not claim the generative edit supplies that guarantee.

For a reusable change-only prompt, read **Single-change preservation edit** in
[community templates](templates-gpt-image-2.5.md). API access and local output
validation remain governed by [models.md](models.md#validation-boundary).
slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack."Stars: 198kUpdated Claude CodeCodexVerifiedInstall
ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read.Stars: 194kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report.Stars: 194kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt.Stars: 194kUpdated Claude CodeCodexVerifiedInstall