Skip to content

veryCoolTimo/imagegen-skills / skills/image-prompt

image-prompt

Flagged

Use when the user wants to turn a short idea into a rich, production-grade image-generation prompt — posters, landing-page or UI mockups, ads, editorial layouts, photoreal scenes, game screenshots, logos, or illustrations. Builds ONE structured, copy-paste-ready prompt optimized for OpenAI GPT Image 2.6 by default, following proven layered-composition patterns (named zones, hex palette, real font references, mood/finish, IP-safety). Triggers on requests like "сделай крутой промт для постера", "нужна картинка/постер/лендинг для…", "generate a prompt for a landing hero", "prompt for an ad / logo / game screenshot".

references/models/universal.md

View on GitHub
# Model adapter — universal (rich natural language)

The model-agnostic default: a single, richly detailed natural-language prompt that works well
across almost any capable model (and is the style of the original gold examples). Use when the
target model is unknown, unsupported, or when the user just wants "a great prompt".

## How to render

Write the full **9-block skeleton as flowing descriptive prose** (see `anatomy.md`) — do NOT
use API-style labeled sections or model flags. One rich paragraph (or a few), covering in order:

1. Concept line (format + medium + fictional brand + reference aesthetic + intended use)
2. Canvas & background (aspect ratio in words + treatment with **hex** + global texture)
3. Named composition zones (position + size% + tilt + content)
4. Subject detail (granular; IP-safe)
5. Typography & text (literal copy **in quotes** + **named real fonts** + color hex + placement)
6. Palette (4–6 **hex** codes, each mapped to an element)
7. Font role map
8. Mood cluster + era/reference anchor
9. Finish/texture + an authority **quality tag** ("Awwwards / Behance-quality … 4k")

Calibrate depth against `gold-examples.md` — this adapter is exactly that level of richness.

## Rules

- Keep everything the gold examples do: real font names, hex for every colour, named zones,
  finish cue, IP-safety ("generic, not resembling any real brand").
- State the aspect ratio in words (e.g. "vertical 2:3", "full-width 16:9") since there are no
  size params.
- No API params, no `--flags` (those belong to the model-specific adapters).

## Output

One clean, self-contained prose prompt in a fenced block, then the assumptions line
(archetype · aspect · brand · model=universal). No CONSTRAINTS/size/quality block (that's
GPT-Image-specific), though you may end with "original design, no logos/trademarks/watermark".

## When to switch

If the user names a supported model, use that adapter instead (gpt-image, midjourney, flux,
ideogram, …). Universal is the safe fallback and the best choice for pasting into an unknown
tool.
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