Skip to content

0xsline/OpenChatCut / src/agent/skills/image-gen

image-gen

Verified

AI image generation via Fal.ai, gpt-image-2, nano-banana, MiniMax image-01, xAI Grok Imagine, and Agnes Image. Use when the user wants to generate or create an image / picture / still.

FileKindSize
SKILL.mdskill md5.6 KB
references/agnes-image.mdmarkdown871 B
references/fal.mdmarkdown955 B
references/gpt-image-2.mdmarkdown2.3 KB
references/grok-imagine.mdmarkdown1.4 KB
references/image-01.mdmarkdown2.2 KB
references/nano-banana.mdmarkdown1.7 KB

references/gpt-image-2.md

View on GitHub
# gpt-image-2 (OpenAI)

Read this document before generating images with `--model gpt-image-2`.

## Capabilities

- Best text rendering among all image models — legible headlines, labels, packaging, UI mockups
- Strongest prompt adherence — follows complex layout and composition instructions precisely
- Max 16 reference images
- Supports `--quality` param: `auto | low | medium | high` (default: `high`)

## Params

| Param       | Values                          | Default | Notes                               |
| ----------- | ------------------------------- | ------- | ----------------------------------- |
| `--quality` | `auto`, `low`, `medium`, `high` | `high`  | Only model with quality control     |
| `--size`    | `1K`, `2K`, `4K`                | `1K`    | 2K/4K are EXPERIMENTAL, slower/heavier |
| `--aspect`  | 10 ratios (see SKILL.md)        | `16:9`  | Computed to exact pixel dimensions  |
| `--input`   | file path or `asset://<id>`     | —       | Up to 16 references                 |

## When to Use

- Text-heavy images: titles, logos, infographics, UI screenshots
- Prompt-sensitive layouts: precise positioning, composition, or style adherence
- Product mockups, packaging, or any design with readable text
- When fidelity to the prompt matters more than matching reference images

## When NOT to Use

- When reference-image fidelity is critical (use `nano-banana` instead)
- When the reference set is larger than the current task needs; prefer fewer, clearer anchors
- When speed matters and quality is secondary (nano-banana is faster for simple prompts)

## Prompt Tips

- Be specific about text content: quote exact strings to render (e.g., `"a poster with the headline 'Summer Sale 2026'"`)
- Describe spatial layout explicitly: "centered title at the top, product image below, price tag in the bottom-right corner"
- For multi-element compositions, describe each element's position relative to others
- Include style descriptors: "flat design", "photorealistic", "watercolor", "3D render"
- Specify background explicitly when it matters: "on a pure white background", "against a gradient from blue to purple"
- For text-on-image edits with `--input`, describe what to change and what to keep: "replace the text on the sign with 'Hello World', keep everything else the same"
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