veryCoolTimo/imagegen-skills / skills/image-prompt
image-prompt
FlaggedUse 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".
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 13.4 KB |
| references/anatomy.md | markdown | 7.5 KB |
| references/archetypes.md | markdown | 8.6 KB |
| references/edit-remix.md | markdown | 6.5 KB |
| references/fonts-palettes.md | markdown | 3.9 KB |
| references/generation.md | markdown | 3.1 KB |
| references/gold-examples.md | markdown | 23.2 KB |
| references/models/flux.md | markdown | 3.8 KB |
| references/models/gemini.md | markdown | 4.4 KB |
| references/models/gpt-image.md | markdown | 10.5 KB |
| references/models/ideogram.md | markdown | 3.3 KB |
| references/models/midjourney.md | markdown | 3.7 KB |
| references/models/recraft.md | markdown | 4.7 KB |
| references/models/reve.md | markdown | 3.9 KB |
| references/models/universal.md | markdown | 2.0 KB |
| references/styles.md | markdown | 5.2 KB |
references/generation.md
View on GitHub# Optional generation (calling image APIs)
The skill is **prompt-only by default**. This is the opt-in path that actually renders a PNG by
calling a provider's image API via `scripts/generate.py` (repo root). Rules:
- **Only generate when the user explicitly asks** ("сгенери", "generate it", "render it") AND a
provider key is present in the environment. Otherwise stay prompt-only.
- **Confirm before spending** — generation costs money (outward-facing, paid). State the
provider/model and that it's a paid call; proceed only on a clear go.
- Keys come from **environment variables only** (never committed, never printed).
- Midjourney has **no official API** (Discord only) — not supported for generation.
## How the skill runs it
1. Build the prompt as usual for the chosen model's adapter.
2. Write the prompt to a temp file (avoids shell-quoting issues), then run:
```
python3 scripts/generate.py --provider <p> --prompt-file <tmp> --out assets/<name>.png [flags]
```
3. The script prints the saved path; then **view the PNG** to confirm it matches, and iterate.
Provider selection: if the user named a model, use its provider; otherwise prefer whichever key
is set (openrouter is the universal one — one key, many models).
## Providers
| provider | env var | endpoint | sync/async | default model | size/aspect flag |
|---|---|---|---|---|---|
| `openai` | `OPENAI_API_KEY` | `/v1/images/generations` | sync | `gpt-image-2.6-sunburst` (`--model gpt-image-2.6-flare`) | `--size WxH` `--quality` `--complexity` `--format` `--background` |
| `openrouter` | `OPENROUTER_API_KEY` | `/api/v1/images` | sync | `google/gemini-3-pro-image` | `--aspect` `--resolution` `--quality` |
| `gemini` | `GEMINI_API_KEY` (or `GOOGLE_API_KEY`) | `…:generateContent` | sync | `gemini-3-pro-image` | `--aspect` `--resolution 1K\|2K\|4K` |
| `ideogram` | `IDEOGRAM_API_KEY` | `/v1/ideogram-v3/generate` (multipart) | sync | `v3` (`--model v4`) | `--aspect` (→ `16x9`) `--style` |
| `recraft` | `RECRAFT_API_KEY` | `/v1/images/generations` | sync | `recraftv3` | `--size WxH` `--style` |
| `bfl` | `BFL_API_KEY` | `/v1/flux-2-pro` | **async (poll)** | `flux-2-pro` (`flux-2-flex` for text) | `--size WxH` or `--aspect` |
| `reve` | `AIMLAPI_KEY` | AI/ML API aggregator | sync | `reve/create-image` | `--aspect` |
## Notes / uncertainties (verify before relying)
- **OpenRouter** is the easiest multi-model path: one key, model slugs like
`openai/gpt-image-2`, `google/gemini-3-pro-image`, `black-forest-labs/flux.2-pro`,
`bytedance-seed/seedream-4.5`.
- **BFL** returns a signed result URL that **expires in ~10 minutes** — the script downloads
immediately.
- **Recraft** V4.1 model string casing is inconsistent in the docs (`recraftv4_1` vs
`recraft-v4.1`); the script defaults to the safe `recraftv3` — pass `--model` to override.
- **Reve** has no confirmed public native API (console is login-gated); the script uses the
**AI/ML API aggregator** (`AIMLAPI_KEY`, model `reve/create-image`). Swap to native once its
spec is confirmed.
- Gemini/gpt-image outputs carry provider watermarks (e.g. SynthID).
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| 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: 322k | Updated | Claude CodeCodex | Verified | Install |
| 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: 198k | Updated | Claude CodeCodex | Verified | Install |
| 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: 198k | Updated | Claude CodeCodex | Verified | Install |
| 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: 198k | Updated | Claude CodeCodex | Verified | Install |