image-generator
替用户优化提示并直接返回渲染结果，降低试错成本。
依赖第三方免费接口，可能不稳定且涉及外部服务调用
通过Pollinations服务将文本描述生成并展示图像。
把用户的文字描述交给 Pollinations AI 生成并直接在对话里展示图片，不需要 API key。用户想要插图、角色头像、场景渲染或任何视觉素材时就触发，用它可以得到实际图像而不是单纯的文字描述。它还会把你的意图改写成更能产出好图的 prompt，减少反复试错。
▸ 展开 SKILL.md 英文原文
Generates images from text descriptions using the Pollinations AI service (no API key required) and displays them directly in the conversation. Use this skill whenever the user asks to create, generate, draw, make, render, or show an image, picture, illustration, artwork, photo, avatar, icon, or any other visual of something — including casual phrasing like "can I get a pic of X", "draw me X", "I need an image for Y", or "what would X look like". Always trigger this even if the user doesn't say the word "image" explicitly but is clearly asking for a picture of something. Do not just describe what an image would look like in words — actually generate and display it with this skill.
帮我安装这个 skill：https://raw.githubusercontent.com/lukan-lawslaf/Claude-Image-Generation-Skill/main/SKILL.mdcurl -fsSL "https://raw.githubusercontent.com/lukan-lawslaf/Claude-Image-Generation-Skill/main/SKILL.md"# Image Generator (Pollinations AI) Turns a user's request into a rendered image using Pollinations AI's free, keyless image endpoint. The core idea: users rarely write great image-generation prompts on the first try, so this skill's job is to translate their intent into a prompt that actually produces a good image, then show the result — not just hand back a URL. ## Workflow 1. **Understand the request.** Figure out the subject, whether the user already specified a style, mood, or composition, and whether they hint at a particular shape/use for the image (portrait, landscape, icon, wallpaper, etc.) — see the sizing guide below. 2. **Enhance the prompt** (see rules below). 3. **Gene