Skip to content

gabelul/pixeltamer-gpt-image-skill

pixeltamer

Flagged

Generate, edit, and compose images with GPT Image 2.6, including transparent PNGs with a real alpha channel. Three backends — OpenAI API key (full feature set), codex CLI on your ChatGPT subscription (agent-driven, no key needed), or the direct ChatGPT image route on the same subscription token (--backend codex-images, PNG-only, opt-in). Four modes — one-shot generate, edit or inpaint an existing image, multi-reference composition (up to 16 inputs blended into one), and multi-image batch with verification. Use when the user asks to make an image, generate a poster, draw a mockup, design an icon or sticker, create an ad creative, build an infographic, blend reference images, edit or inpaint an existing image, or produce a cutout asset with a transparent background — "transparent PNG", "no background", "alpha channel", "works on any background". Not for: stripping the background off an existing image you want to keep (that's rembg or ImageMagick, not a generator); file operations that don't change the picture (compression, format conversion, resizing); reading, describing or OCRing an image; video.

examples/README.md

View on GitHub
# Examples

A small curated set of generations to give the eye a sense of what's possible. Each one was made by one of the source repos this skill consolidates — the originals were generated against gpt-image-2 with the prompts shown below.

The shipped set is intentionally small — about 3 MB total after pngquant compression. Running pixeltamer with these prompts will reproduce them at any size.

## product-photo.png

Square minimalist product photograph. Demonstrates editorial product styling on a cup-of-coffee subject — the kind of shot you'd put above the fold on a product page.

```
Elegant minimalist product photography: a single white ceramic coffee cup
on dark marble surface, steam rising softly, dramatic side lighting.
```

- Size: 1024×1024
- Recipe: `recipes/product-photo.md`
- Backend: codex (in the source repo); reproduces fine on API too

## editorial-poster.png

Vertical poster composition with a single hero subject and atmospheric backdrop. Good demonstration of mood-driven editorial work.

```
A lone astronaut standing on the edge of a crater on Mars, looking at
Earth rising on the horizon, cinematic composition.
```

- Size: 1024×1536
- Recipe: blends `recipes/editorial-cover.md` with cinematic lighting from `references/prompting.md`

## landscape-cinematic.png

Wide cinematic landscape — Mediterranean coastal city at night. Useful as a reference for atmospheric night scenes with practical lighting (windows, lamps).

```
A breathtaking coastal Mediterranean city at night, white-washed
buildings cascading down hillside.
```

- Size: 1536×1024
- Recipe: ad-hoc; references `references/prompting.md` lighting + composition vocab

## infographic-llm-stack.png

Tech infographic with a 5×6 grid of brand logos. Demonstrates gpt-image-2's grasp of real product visuals (logos render correctly because the model has them as references in its training).

```
Tech infographic: "THE AI STACK — 2026" title + 30-logo grid (5 rows × 6
columns) showing frontier labs, models, image/audio, dev tools, and
infra companies. All real logos rendered accurately.
```

- Size: 3840×2160
- Recipe: `recipes/infographic.md`
- Backend: API (multi-logo accuracy benefits from `--quality high` and reliable text rendering)

## Reproducing these

Each prompt above can be passed straight to `pixeltamer generate`. The exact pixel output won't be identical (image generation has stochasticity), but the composition and style will be in the same neighborhood.

```bash
pixeltamer generate -p "Elegant minimalist product photography..." \
  --size 1024x1024 --quality high -o my-product-photo.png
```

For the multi-reference recipes (compose mode), see `recipes/meta-ad.md` and `recipes/product-photo.md` for worked examples that need real reference images you'd supply yourself.
brainstormingobra/superpowersYou MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, or planning anything new, in software or out of it (a talk, a business, a renovation).Stars: 322kUpdated Claude CodeCodexVerifiedInstall
brand-guidelinesanthropics/skillsApplies Anthropic's official brand colors and typography to any sort of artifact that may benefit from having Anthropic's look-and-feel. Use it when brand colors or style guidelines, visual formatting, or company design standards apply.Stars: 198kUpdated Claude CodeCodexVerifiedInstall
frontend-designanthropics/skillsGuidance for distinctive, intentional visual design when building new UI or reshaping an existing one. Helps with aesthetic direction, typography, and making choices that don't read as templated defaults.Stars: 198kUpdated Claude CodeCodexVerifiedInstall
mcp-builderanthropics/skillsGuide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK).Stars: 198kUpdated Claude CodeCodexVerifiedInstall