Image

根据儿童年龄与共读表现，评估适合的陪伴内容、建议照片素材，引导选择单页互动教辅或连续故事，确认梗概与画风后生成双语绘本并检查故事逻辑、跨页面容和图文一致性。适用于定制亲子绘本、宝宝认知互动图、照片角色绘本，以及这些作品的局部修改和质检。

Use when preparing photos for upload to Adobe Stock or Shutterstock contributor accounts, when a stock photo needs a title, description, category, or keywords, or when a batch of stock images needs a metadata CSV

Pixel HUD UI, a complete UI framework in a retro flight-computer / game-HUD style (pixel CJK font, square 1px-ruled panels with sky title-bar strips, CRT scanlines, intent-colored solid keys, lamp switches, segmented gauges), with a design spec for every control (sizes, states, usage, accessibility) and a drop-in CSS kit. Use it to design or build a new interface in this style, or to restyle an existing web app (plain HTML, React, Next.js, Vue, Svelte, Tailwind, shadcn/ui) into it. Use it whenever the user wants a pixel-art, 8-bit, retro game HUD, cockpit, terminal or sci-fi console look, a "game UI" for a dashboard or tool, or mentions 像素风, 像素 HUD, 复古终端, 游戏 HUD 风格, 飞行计算机界面, even when they…

铸想 ForgeForm（ZYKT studio 出品）—— 用一句中文描述生成可打印的 3D 模型。当用户想「文字/图片 → 3D 模型 → STL」「批量出图」「给本地 AI 工作流换生图/建模/视频模型」时使用。已装好的机器上直接调；没装的机器按本文安一遍即可。

摸鱼PPT（moyu-ppt）：把文章、演讲稿、路演或汇报材料做成视觉主导的图像版 PPT/PowerPoint：先确认需求与整体感觉，再做灵感检索、字体与风格系统、全篇空间定位图、逐页生图，交付一图一页的 pptx。也处理只要其中一个阶段（找灵感、定风格字体、只出方案）、修改已有图像版 PPT 的指定页，以及保存、查找、复用用户收藏的 PPT 风格模板；支持中文标题字形参考图。公众号文章、Excel 数据可作为素材。独立知识卡片/单页图文、原生可编辑幻灯片或修改可编辑 PPTX、HTML 演示不属于本技能；只解释、翻译内容或与 PPT 无关的生图不启动本技能。

Builds a console script with embedded prompts that repeats image generation on Higgsfield. Use it for text-only generation, several variants of the same original, swapping reference images in numbered order, and sorting a reference folder. Supports general image work such as characters, logos, products, backgrounds and spaces; video generation is excluded.

為 0–6 歲小朋友製作個人化繪本：開工前逐條問清（每條俾選項）小朋友幾歲、男仔定女仔、單語／雙語／三語、一行幾多隻字、總共幾多頁、揀邊種畫風，同埋拎家庭相做人物參考。全程只有兩個確認位（一次過批設定＋全本文字、一次過批封面），批完封面就一口氣做晒成本書，唔逐頁停低問。內文用加大字體版面（中文 24pt／英文 23pt 起，舊版係 19／13.5）。開工前必須先輸出「讀檔回報」證明真係讀過本檔，讀唔到要老實講明，唔准靠估扮讀過。觸發：整繪本、picture book、故事書、幫小朋友整本書、戒尿片書、返學書、睡前故事書

将产品观点、AI/技术原理、系统架构、流程、对比或抽象概念，制作成具有 Apple Keynote 发布会质感的中文/英文解释型配图。适用于用户说“用之前的 Keynote 风格画图”“给这段文字配图”“画一个架构图”“做简洁易懂的概念可视化”“不要文字”时；强调浅色高留白、精致柔和 3D、克制配色与一眼看懂的关系。非摄影写实、非复制 Apple 官方素材，且不替代严格数据可视化。

Generate a WeChat pictorial lead-character turnaround sheet (主角三视图/定妆图) from a REQUIRED user-uploaded person photo: front, three-quarter, side, half-body portrait, plus character-lock.md. Enforces role-correct wardrobe. Trigger on 主角三视图, 角色定妆图, character sheet, turnaround, 上传人物照片生成三视图, character-lead-sheet. Do not invent a face without a person reference image.

链接转深度研究、2000px 矢量高清信息图与推特 (X) 爆款文案全自动生成工作流。支持网页/推特文章穿透研读、4 维背景拓研、双重事实核验门禁、真实 SVG 手绘架构流、2x Retina 无损渲染与全套推特发文文案包。

DanLogo · Soft Order 柔性秩序：有机的形，机械的精度——几个柔软的部分按一套看不见的规则组成一个正在发生的过程，只让一处关系打开，成为全部焦点； 以《周易》“生生”、谢赫“气韵生动”、达西·汤普森《生长与形态》、有机抽象与有机现代、程序化设计为思想来源；字从零绘制、软形硬字，色以物名、只亮一处。 按九个阶段推进：简报 → 概念（每轮 5 套）→ 原型（每个概念 ≥3 套）→ 图形精修 → 定稿（定制字 + 配色）→ 体系 → 场景 → 提案 → 总结。 Use this skill when the user wants to design, redesign, refine, extend or present a logo, wordmark, 标志, 字标, 中英文组合标, 品牌视觉, VI or brand identity in an organic, quietly precise spirit — 柔性秩序, 有机, 生长, 有机几何, 灵动, 不规则中的规则, 隐性秩序, 温润, 去 AI 味, 去拼搭感, organic, growth, hidden order — when they name danlogo-softorder, or when they continue a project whose decisions.md names it. For a logo request that names no style, use it unless another installed danlogo style fits…

DanLogo · Geometric Modular 几何模块：先设计一套同源、同度的几何构件，再按看得见的规则把它们组合成一个完整的图形——标志是最有意义的一个组合，整个体系是同一套构件的其他组合。 按九个阶段推进：简报 → 概念（每轮 5 套）→ 原型（每个概念 ≥3 套）→ 图形精修 → 定稿（定制字 + 配色）→ 体系 → 场景 → 提案 → 总结。 Use this skill when the user wants to design, redesign, refine, extend or present a logo, wordmark, 标志, 字标, 中英文组合标, 品牌视觉, VI or brand identity in a geometric, modular, systematic spirit — 几何, 模块, 模块化, 构件, 拼搭, 网格, 系统化, Bauhaus, modular system, kit of parts, grid, geometric identity — when they name danlogo-geomodular, or when they continue a project whose decisions.md names it. For a logo request that names no style, use it unless another installed danlogo style fits better; if several danlogo styles are…

DanLogo · Concrete 建筑感：像安藤忠雄那样做标志——一块沉默的几何实体，被一次有分寸的减法打开，让光进来；颜色就是材料本身。以安藤为核心、以极简建筑谱系为范围（康、巴拉甘、卒姆托等）；四种形式语言：体块·虚实、分面·光影、圆弧·层叠、模块·秩序。不限行业，按审美选用。 按九个阶段推进：简报 → 概念（每轮 5 套）→ 原型（每个概念 ≥3 套）→ 图形精修 → 定稿（定制字 + 配色）→ 体系 → 场景 → 提案 → 总结。 Use this skill when the user wants to design, redesign, refine, extend or present a logo, wordmark, 标志, 字标, 中英文组合标, 品牌视觉, VI or brand identity in an architectural, Tadao Ando-like minimalist spirit — 建筑感, 安藤忠雄, 极简建筑, 清水混凝土, 光, 体量, 几何, architectural, concrete, minimal architecture, tectonic, geometric — when they name danlogo-concrete, or when they continue a project whose decisions.md names it. For a logo request that names no style, use it unless another installed…

DanLogo · Swiss Tech 静几何：以字为标——中英文字形从零构造到可以当图形看，再从字标里取一处或几处组合、改造成符号（字成一体，取字成符）； 修正几何（圆与方之间的曲线）、模数与基准、大静小响的色彩分离、墨阱式的修正细节，带来现代、潮流、科技的气质。 按九个阶段推进：简报 → 概念（每轮 5 套）→ 原型（每个概念 ≥3 套）→ 图形精修 → 定稿（定制字 + 配色）→ 体系 → 场景 → 提案 → 总结。 Use this skill when the user wants to design, redesign, refine, extend or present a logo, wordmark, 标志, 字标, 中英文组合标, 品牌视觉, VI or brand identity led by custom lettering in a Swiss / rational / precise spirit — 字体设计, 定制字体, 字标设计, 字母标, 瑞士风格, 理性, 精密, 科技感, 潮流, 硅谷大厂气质, custom type, logotype, lettermark, Swiss style, modular, superellipse — when they name danlogo-swisstech, or when they continue a project whose decisions.md names it. For a logo request that names no style, use it unless…

从参考图片或人物到目标造型的示例中反推可迁移的视觉规则，编写通用生图提示词，并在用户要求时按实际结果迭代。适用于画风反推、角色转换模板、纯提示词版本和跨角色复用；不是直接恢复原始提示词或默认启动付费生图。

Optimize, rewrite, and craft image generation and editing prompts tailored specifically for Alibaba's Qwen-Image-2.1 diffusion model. Use this skill whenever the user wants to generate images with Qwen 2.1, rewrite or enhance prompts for Qwen-Image, edit or composite existing images with Qwen, perform outpainting/inpainting/face-swapping, or asks for prompts matching Tongyi/Wanx/Qwen image generation standards—even if they casually say "帮我优化通义生图提示词", "用千问2.1出图", or "Qwen改图".

给任何故事里的角色真出参考图（小说改编、自己原创的故事、单独设计一个角色都行，不需要小说原文）： 一段话描述角色，拆成分层字段、补全后确认， 先出一张正面全身锚点，其余视图（大头照、90° 侧面、背面、细节、45° 大头照）都只参考这张锚点， 按需分档出图。每张图带标识、可单独重出，重出后自动标出哪些图过期。 支持 Qwen Image（ComfyUI）、codex 内置出图、OpenAI Images API（GPT Image 2）和自定义命令，主图与派生图可以使用不同模型。 产出 asset.json + 一组 PNG + 双击就能开的报告。零依赖。 只要提示词、不出图的话使用 novel-characters。 Use when asked to 出角色图、做角色参考图、给我的角色出图、定妆照、真出三视图、character reference images。

Create or revise game-video thumbnails from gameplay, game-art references or an existing thumbnail. Use for thumbnail composition, text removal and art-style corrections; not video editing or general image generation.

生成「每日一句 / 打卡」风格社交卡片图，适用于朋友圈、公众号、小红书。当用户要求做卡片图、日签、打卡图、语录海报、带标题+副标题+日期的分享图，或要调整主题配色/贴纸元素、从背景图自动取配色时使用。触发词：每日一句、打卡图、日签、分享图、朋友圈配图、小红书封面、公众号首图、卡片生成、daily card、社交卡片。

用内置 ImageGen 构思、生成、重设计或局部修改海报、品牌分享图、封面和宣传图。适用于需要整张视觉出图、保护品牌参考与真实素材、逐项验收和继续改稿的 ChatGPT Work 或 Codex 任务。

Turns any grey 3D scene (a .blend with or without a camera, a glTF/GLB, OBJ, FBX, USD, PLY or STL file, a JSON scene spec of primitives, or an idea written as a spec first) into the same scene with ordinary PBR materials and fitted lights that re-render like a photo from its camera: a photoreal target of the fixed view (Nano Banana 2, conditioning ladder checked against the object-id mask), albedo, roughness, normals and metric depth from the target, back-projected into each object's own UV textures as Principled BSDF materials, lights fitted through Cycles basis renders, then a measured loop (LPIPS, per-object errors, bump detail, a fresh reviewer's list applied as geometry edits) and a…

从参考图提取黑白线稿、生成高斯模糊色彩参考、反推中文提示词，再通过多参考重建提升图片视觉质量。用于保形重绘、线稿上色和写实质感重建；保持原图身份、比例、姿态及色彩，优先选用模型最大原生尺寸，支持原生4K或标注来源的4K放大交付。不用于视频生成或证据修复。

Turns one reference photo of a subject (an animal, a character, an object) into editable low-poly 3D meshes under hard triangle budgets (for example 100, 200, 500 and 1000), each matched to the photo through one locked camera: fixed targets made once (mask, landmarks, image windows, all hashed), a camera fitted jointly with a parametric proxy that supplies the inferred hidden side, a build script per budget that allocates triangles by region and welds every part into one closed surface, a shared seeded vertex optimiser that never changes the count, a four-way triangle count (mesh file, evaluated .blend, re-imported GLB, parsed GLB), official metrics from a Blender render (silhouette IoU…

실제 사진처럼 보이는(photorealistic) 이미지 생성 프롬프트를 사용자와 단계별로 함께 만드는 스킬. 피사체 → 환경 → 동작과 끝점 → 카메라 → 조명과 스타일 순서로 AskUserQuestion 객관식 질문을 던지며 프롬프트를 쌓아 올리고, 최종적으로 영문 프롬프트와 한글 번역을 함께 낸다. 사용자가 "이미지 만들어줘", "사진 프롬프트", "실사 이미지", "현실적인 장면", "AI 이미지 프롬프트 다듬어줘", "Midjourney/GPT Image/Nano Banana/Imagen/Flux용 프롬프트" 등 이미지 생성이나 이미지 프롬프트를 언급하면, 스타일을 명시하지 않았더라도 이 스킬을 사용한다. 일러스트·로고·다이어그램처럼 명백히 사진이 아닌 결과물에는 쓰지 않는다.