← Back to list
🎨
imagegen Checked
生成网站与内容所需的一切视觉素材——用户说 生成图片、配图、插图、画一张、出一套图、logo、吉祥物、封面、海报、og 图、分享图、favicon 源图、用户场景图、真人图("一个人坐在电脑前"…
🎨 Design & Creative By yan-labs Version v1.0.0 Updated 2026-09-11
8.9Overall rating

Installation

🤖 Install via AI

Copy the prompt below and send it to your AI assistant (e.g. Claude Code) — it will follow the instructions and install automatically.

Install the "imagegen" skill by following the instructions at https://skill123.me/install/imagegen.
⌨️ Command line install

Run in your terminal — downloads and installs to ~/.claude/skills/.

curl -fsSL https://skill123.me/install/imagegen.sh | bash
📦 Download ZIP

Download the zip and extract it into your skills directory (e.g. ~/.claude/skills/), then restart your session.

⬇ Download v1.0.0 · 10 KB

About this skill

Overview

生成网站与内容所需的一切视觉素材——用户说 生成图片、配图、插图、画一张、出一套图、logo、吉祥物、封面、海报、og 图、分享图、favicon 源图、用户场景图、真人图("一个人坐在电脑前")、手绘/蜡笔/水彩风插画、产品宣传图、电影感画面、"网站需要配图"、image gen、imagegen 时必用。也覆盖 rankup 建站流程里的"页面缺图、还是占位图、og:image 没图、每页要独立分享图、favicon 还没做"——这些场景不许用占位图,必须真实生成,一律加载本 Skill。底层走 Codex 内置的 OpenAI 图像生成工具;本 Skill 负责把需求翻成好提示词、跑通、验收、压缩、落盘。

Source

  • Repo: https://github.com/yan-labs/yan-skills
  • Path: imagegen

Score breakdown

Trigger
7.5
Rich Chinese trigger vocabulary covering many asset types plus rankup no-placeholder scenarios, but no When-Not and English variants are only image gen and imagegen.
Structure
7.3
Tight 183-line SKILL.md with template tables; README.md ships and rankup reference paths are external.
Workflow
9.6
Strong generator design: prompt recipe with escape hatch, use-case templates, three-check acceptance, compression pipeline, curl-200 post-wiring verification.
Content
10.0
Exceptionally concrete: measured file sizes, timings and quantified acceptance thresholds throughout, imperative style.
Engineering
9.0
Valid frontmatter, no scripts; references to rankup SKILL.md and seo-growth.md resolve only when the sibling skill is installed.
Security
10.0
No scripts, no secrets, no injection; sandbox flag handling is conditional on machine authorization with mandatory disclosure.
Focused, well-measured image generation skill with a six-part prompt recipe, per-use templates, three-check acceptance and verified end-to-end run record; flaws are a bundled README, Chinese-only description, and rankup cross-references not in the package.