Image

通过弹窗引导新手确认屏幕、行业设备及制作范围，制作可选布局色调的工业HMI预览，再展开页面；支持仅调用技能启动、局部编辑及MCGS原生复制，无需MCP或外部供体。

Texture existing Blender game models with palette-strip gradient UV mapping from a concept image. Use only when the user explicitly invokes /tb-hoidz, $tb-hoidz or tb-hoidz. Do not trigger for generic requests such as gradient mapping, UV gradient or tô màu model bằng palette unless tb-hoidz is named, so it does not collide with other team members' texturing skills.

Skill tạo hình minh họa đẳng trục (isometric) tương tác để giải thích một ý tưởng kỹ thuật: agent "phát minh" một vật thể cơ khí mà cơ chế hoạt động của nó chính là khái niệm cần giải thích, dựng hàng chục chi tiết nhỏ bằng SVG + WebGL (tùy chọn 3D beta), đóng khung thẻ có tiêu đề, số liệu trực tiếp và chuyển động mượt theo con trỏ/bàn phím, rồi tự kiểm tra hình bằng audit hình học và ảnh chụp headless Chrome. Xuất HTML độc lập, SVG đơn hoặc component React/Next.js.

Image optimization analysis for SEO and performance. Checks alt text, file sizes, formats, responsive images, lazy loading, CLS prevention, image SERP rankings (via DataForSEO), and image file optimization (WebP/AVIF conversion, IPTC/XMP metadata injection). Use when user says "image optimization", "alt text", "image SEO", "image size", "image audit", "optimize images", "image metadata", "image SERP", "convert to webp", or "image file optimize".

Buat atau sunting ilustrasi raster dengan imagegen GPT lewat Codex CLI yang login memakai langganan ChatGPT, tanpa API key. Pakai saat pengguna meminta gambar, ilustrasi, atau infografis bergambar dibuat dengan "imagegen GPT", "gambar dari GPT", "Codex imagegen", atau "/claude-imagegen", terutama dari Claude Code. Skill menulis brief, menjalankan codex exec dengan gambar rujukan gaya, memeriksa hasil secara visual, lalu menyimpan PNG asli, WebP opsional, dan arsip brief. Jangan dipakai untuk diagram yang lebih tepat dibuat sebagai SVG, HTML, atau kode.

核对 AI 图与实拍中的商品差异，逐图逐人展示编号放大对比；用户指定修改项后，自动选择合适方法局部修复并展示修后复核。适用于服装、鞋帽、箱包等商品；人物一致性仅在用户明确要求时检查，不用于背景评价或像素差分。

Convertit un véhicule FiveM / GTA V (zip ou dossier de ressource avec .yft/.ytd/.meta/.awc, dlc.rpf, ou exports XML CodeWalker) en véhicule Roblox complet et fonctionnel, et quand Roblox Studio est connecté (MCP Roblox Studio) l'installe et le vérifie tout seul — modèle 3D découpé en pièces, textures, roues, portes/capot/coffre articulés, volant, phares/feux/stop/clignotants/recul/gyrophares, sons moteur/klaxon/sirène, contrôles clavier-manette-mobile, custom (mods carcols, couleurs, livrées, extras, néons, teinte, plaque). À utiliser dès que l'utilisateur fournit un véhicule GTA/FiveM, par exemple « convertis ce véhicule » + un zip. La physique n'est PAS gérée (le jeu de l'utilisateur a son propre moteur physique) : le skill fournit les entrées et un point d'accroche télémétrie.

Generates or edits raster images (illustrations, photos, icons, textures, mockups, posters, logo concepts, transparent cutouts, placeholder assets) by driving the codex CLI's built-in image tool, and saves them to a path. Use by default whenever an image is to be generated or an existing one changed ("make the sky purple"), and when the user names Codex, ChatGPT or OpenAI image generation. Runs on the Codex login's quota, with no API key. For Google/Gemini image generation with an API key, use google-image-gen instead.

电商详情页套版技能，从 PSD/PSB 母版为每款生成独立可编辑长图，保留可替换的嵌入式智能对象及可编辑文案和表格。按需输出 790px JPG 模块。

出图-白底透明图V2：生成800×800白底JPG、透明PNG及按款汇总的高清像素层与独立蒙版PSD，保留未缩放原始素材，支持边缘自检、局部修复、确认锁定和“开始PS精修”口令执行已排队的PS任务。用于产品或模特素材去背景及其精修队列；普通尺寸调整不启用。

Apply a brand's defined visual system to any fixed-format piece that isn't a website — decks and presentations, one-pagers and sales sheets, social posts and graphics, blog/OG thumbnails, business cards, brochures, flyers, posters, banners, billboards, and paid ads. Use when the user mentions 'design a deck,' 'pitch deck,' 'sales deck,' 'presentation,' 'slides,' 'one-pager,' 'sales sheet,' 'leave-behind,' 'social post,' 'social graphic,' 'blog thumbnail,' 'OG image,' 'business card,' 'brochure,' 'flyer,' 'poster,' 'banner,' 'billboard,' 'print,' 'ad creative,' 'display ad,' 'collateral,' or wants a branded document or graphic that isn't a website. Mostly screen-first; print is one destination among many. Ships as a true-size viewer plus real exports.

Idea-to-printable-3D-model studio for FDM printers. Use when asked to "model / design / make me a <thing> to print", "3D model this", "CAD this", "custom mount/case/holder/adapter/lid/bracket", "fidget", "print-in-place", "something that moves", "flexi <animal>", "figure / statue / mascot of <x>", "text to 3D", "gridfinity", "insert for my <thing>", "split this for my bed", "will this print", "calibrate my printer", "see it in AR", "put my name on", "add a keychain loop", "add magnets", "fix thin walls", "scale to my head", "make me a box", "keychains", "sign", "phone stand", "lithophane", "max model", "/max-model", or "forge". Code-CAD (build123d/OpenSCAD/bpy) + MechKit verified mechanisms + local text-to-3D + flexify + preflight failure prediction; hands a verified STL to your slicer.

Grilling with diagrams — each round of the interview is also rendered as a page where every structural decision shows one diagram per option. Use when grilling a design whose choices shape the architecture (data model, ownership of a write, a seam, a lifecycle), or when the user asks for schemas or a visual grill.

Design physical or 3D-printable objects — industrial products, consumer devices, robots, enclosures, furniture, vehicles, wearables, helmets, tools, and packaging structure. Use whenever proportion, packaging of internal components, ergonomics, surfacing, CMF (colour/material/finish), manufacturability, assembly, or 3D printing matters, including "design a case for this PCB", "make a printable robot", or "develop form concepts for our device". Do not use when the deliverable is only a flat picture of an object (image-design).

Render 3D models/prefabs in a Unity (URP) project into transparent, outlined sprite icons for inventories, shops, item cards and grid backpacks - sized to the item's cell footprint, auto-posed so long weapons lie diagonally and compact props fill the slot. Use when a project needs item/inventory icons from its 3D art, has placeholder or missing icons, wants icons regenerated after art changes, or wants Backpack Battles-style grid item art.

Monta apostilas, resumos, mapas, revisões, questões e flashcards ilustrados de qualquer assunto (faculdade, medicina, concursos, escola, certificações) em HTML/PDF A4 com design editorial, figuras de acervo aberto com crédito (OpenStax, Commons) ou geradas a partir de dados, estilos de cor (sóbrio, colorido, pastel, P&B) e material no idioma do usuário. Use para /resumo, /esquema, /lei-seca, /comparar, /pegadinhas, /questoes, /flashcards, /mnemonico, /edital, /cronograma, /revisao, /apostila, /validar, /grifo ou qualquer pedido de material de estudo, em qualquer língua (study guide, summary, cheat sheet, flashcards, resumen, apuntes, fiche de révision).

Turn a described business process, workflow, or domain into one or more Domain Stories (domainstorytelling.org) rendered as pictures, and as Egon.io files that open at egon.io. Use when asked to draw, model, visualise, or map a process as a domain story, to run a Domain Storytelling workshop, or to make sense of who does what with what in a domain. Also use before designing software when the business process behind it has never been written down.

1219 ready-to-run matplotlib generators for publication-quality scientific figures, on an enforced Times New Roman design system with a bundled metric-compatible fallback font and a typography linter that reads the fonts actually embedded in the rendered PDF. Covers core statistical charts (threshold regression, raincloud, ridgeline, clustermap, radar, Sankey, treemap, UpSet, Taylor, volcano, Manhattan, forest, 3x2 six-panel grids), geospatial mapping, ML and SHAP interpretation, bio-omics, climate, graphical abstracts, physics and engineering, economics, clinical trials, chemistry and materials, Bayesian/MCMC diagnostics, neuroscience and signals, optimisation and metaheuristics, UAV and robot path planning, remote sensing, power systems, transport, hydrology and ocean, ecology, astronomy, manufacturing quality, computer vision, network architecture diagrams, attention and feature visualisation, 3D vision and point clouds, generative models, training and benchmark

Make game-ready 2D characters, creatures, props, icons and FX as master stills, sprite sheets or animation clips. Plan size, motion and finish; generate the art (image generation is the default, through generate2dmedia route_media.py - the API when a key is configured, else the local Codex or Grok CLI); approve one master still per character; then key, slice, register, scale, palette, QA and export frames for a game engine (Aseprite JSON, Godot SpriteFrames or Sprite3D). HD finish by default, pixel finish on request. Character animation is one image-to-video clip per action from the master (video2dsprite); sheets are for FX, icons and props. Use for sprites, masters, sheets, frame-by-frame animation and packaging frames from any source. Not for maps, tiles or backgrounds (generate2dmap), video clips (video2dsprite), code-drawn art (codeart2d, only on request) or calling an image or video API by itself (generate2dmedia).

Design, draw, and verify figures for security/systems papers (IEEE S&P, USENIX Security, CCS, NDSS, journals, IEEE S&P Magazine), in the style of strong top-4 papers. Use for 画图/作图/图表设计/论文配图, choosing the right chart for a claim, making a figure print-safe, writing captions that state the takeaway, or converting a matplotlib figure into a camera-ready vector PDF. Ships runnable examples with synthetic data (examples/), a matplotlib style module (scripts/figure_style.py), a generator (scripts/make_examples.py), and a pre-submission checker (scripts/check_figure.py).

Configure a custom OpenAI-compatible image service with a local setup page, or generate and edit images through that service with support for the gpt-image model family and automatic model selection. Use for codex-image2 setup, API configuration, text-to-image, image edits, and batch image generation.

Design and refine SVG logos, icons, brand assets and visual graphics for websites and user interfaces. Use for visual identity, icon systems, decorative or animated web artwork and vector refinement. Explanatory diagrams belong to Editorial SVG Diagrams; full page layouts belong to Frontend Design.

Generate publication-quality academic illustrations through a local Codex app-server bridge that uses Codex native image generation. This is a separate experimental alternative to `paper-illustration`, intended for Claude Code users who want a GPT-image-style renderer without modifying the original skill.

Making a picture and placing it in the project: art style, image brief, generation prompt, seed, run the generation, edit a region, upscale, file format, srcset. Use when pixels must exist or change.