Image

ascii-art-hd

github.com

Turn a picture, animated GIF, frame folder or 3D .obj mesh into high-resolution, per-character-colour ASCII art (still or animated/rotating), with HTML, PNG, ANSI, plain-text and ASCII Mate .ascii.json output. Use when the user wants detailed ASCII art of a subject (knight, horse, animal, logo, portrait, 3D model), a rotating ASCII turntable, a colour ASCII version of an image, or asks why their hand-built low-resolution ASCII/3D art is unrecognisable.

$ npx skills add naytngard/ascii-art-hd
Skill anzeigen→

onboarding-paywall-designs

github.com

Designs high-converting, honest onboarding and subscription paywall screens for an existing app and renders them as PNG images (iPhone/Android, light/dark, A/B variants). Reads the project first, writes a project brief and a conversion strategy, asks only the questions the code cannot answer, then renders HTML/CSS templates with Playwright and runs automated quality checks (overflow, WCAG AA contrast, CTA above the fold, mandatory store elements). Use when the user asks for paywall designs, onboarding screens, subscription screen mockups, trial/pricing screen images or paywall A/B variants. Produces images and strategy only, never app code.

$ npx skills add gollodev/onboarding-paywall-designs
Skill anzeigen→

long-image-to-x

github.com

Slice one long image (an illustrated essay, webcomic, infographic, long screenshot, or paged PDF export) into a set of images ready to post on X (Twitter). Trims side margins, finds cut points between content blocks, writes each set to its own folder, optionally stamps a watermark, checks the result visually, and gives posting advice. Use this whenever the user wants to split, cut, crop or slice a long image, long-form post, comic or document into several images for X or Twitter, asks for a four-image post, or says something like "make this fit X", even if they never say "slice" or "skill". Produces the images and a plan only; it does not post anything.

$ npx skills add MacchaCat/long-image-to-x
Skill anzeigen→

seedream-55

github.com

使用 Nange AI Seedream 5.5 API 生成图片,支持文生图、图生图(最多 10 张参考图融合)、透明背景(贴纸/抠图)、图层拆分(把海报拆成底图 + 透明图层,可选文字 OCR),1K / 1.5K / 2K 三档分辨率或精确像素,按输出张数 + 参考图张数计费。当用户说"seedream"、"用 seedream 画"、"即梦"、"豆包生图"、"透明背景"、"贴纸"、"抠图"、"图层拆分"、"拆图层"、"把海报拆成图层"、"多图融合"、"参考这几张图"、"生图"、"画图"、"生成图片"、"帮我画"、"把这张图改成"等涉及 AI 图片生成、透明背景图、图层拆分或图片编辑的请求时触发此技能。首次使用会引导配置 API Key,后续直接生成。

$ npx skills add xiaonan0527/seedream55-gen
Skill anzeigen→

brainlab-gercekci-kusurlar

github.com

Video ve image (fotoğraf) prompt'lara gerçekçi kamera, lens, sensör, medya ve operatör kusurları ekler; sahneyi daha doğal, daha "gerçekten çekilmiş" gösterir. Üç şekilde çalışır. (1) Sıfırdan, bir sahne ya da fikirden kusurlu detaylı video prompt yazar. (2) Başka bir video prompt skill'inin (sd25-pe, brainlab-cinematic-output, brainlab-realistic-output, brainlab-timed-video-prompt, seedance-prompter vb.) çıktısını analiz edip aralara önemli kusurları yerleştirir. (3) Kullanıcının yapıştırdığı hazır bir video prompt'a eksik kusurları ekler. Her seferinde, eklemeden önce hangi kusurların eklenmesini istediğini tek tek sorar. Ses kusurlarını sadece kullanıcı isterse ekler. MUTLAKA tetikle. "/brainlab-gercekci-kusurlar", "gerçekçi kusurlar", "kusur ekle", "daha gerçekçi yap", "handheld hissi ver", "daha doğal görünsün", "camcorder hissi", "found footage", "bu prompt'a kusur ekle", "odak kayması ekle", "kamera titremesi ekle", "AI hissi gitsin", "fazla temiz görünüyor", "bu çok kritik",

$ npx skills add ecinaro/brainlab-realistic-output
Skill anzeigen→

jisui

github.com

Jisui (自炊) processing of scanned page images into a clean ZIP, CBZ or PDF — deskew (rotation + sheet-feed shear), minimal margins with one uniform page size, B&W binarization with pictures kept grayscale, front/back cover and jacket handling (include, leave out, or save as separate files; from a given file, the scans, the front/back of the jacket, or the web), OCR text layer (ocrmypdf + Apple OCR), PDF bookmarks from the table of contents, and Title/Author metadata editing without recompression. Use this whenever the user has a folder of scanned images (from a ScanSnap or other scanner) of a book, magazine, manual, notes, or any document and wants them turned into a ZIP/CBZ/PDF, cleaned up, deskewed, cropped, binarized, OCR'd or bookmarked — even if they don't say "jisui", e.g. "turn these scans into a PDF", "make a PDF from my scanned images", "OCR this scanned PDF", "add a TOC to this scan".

$ npx skills add 0stellensatz/skill-jisui
Skill anzeigen→

aard-visual-director

github.com

Creates original, art-directed image concepts and production-ready prompts across portrait, fashion, editorial, cinematic, surreal, fantasy, futuristic, historical, product, and environmental imagery. Use when a user asks for image ideas, visual concepts, art direction, styling, reference-image guidance, or prompts for image-generation tools.

$ npx skills add aard-anonthanat/aard-visual-director
Skill anzeigen→

optic

github.com

Use this skill when generating, prompting, editing, or quality-checking photorealistic or authentically crafted visuals (portraits, lifestyle scenes, commercial products, architecture, culinary, editorial) where physical believability is essential. Do NOT use for technical diagrams, charts, UI mockups, flat vector logos, or wireframes.

$ npx skills add Zevnex/optic
Skill anzeigen→

brainlab-realistic-output

github.com

brainlab-cinematic-output'un gerçekçi versiyonu. Sinema kadrajı yerine "birinin gerçekten çektiği an" hissi veren image (görsel) prompt'ları üretir. Sadece görsel skill'dir, video ve ses yoktur. Ağırlık dört şeyde. (1) Kişisel insan detayı (ten, saç, el, kıyafet aşınması, mikro ifade) ve nesne detayı, (2) gerçek ve motive edilmiş ışık, (3) gerçek kamera, varsayılan iPhone ama bir kamera bankasından seçilir, (4) opsiyonel ek olarak bağlı alt-skill brainlab-gercekci-kusurlar (kamera, lens ve insan kusurları). İki mod. Metinden prompt'a ve görselden prompt'a (tersine mühendislik). MUTLAKA tetikle. "/brainlab-realistic-output", "realistic output", "gerçekçi çıktı", "gerçekçi prompt", "iPhone ile çekilmiş gibi", "telefon kamerasıyla çekilmiş gibi", "candid", "gerçek fotoğraf gibi", "AI gibi durmasın", "plastik durmuyor mu", "daha gerçekçi görsel", "günlük hayat fotoğrafı", "UGC tarzı", "belgesel gerçekliği", "cinematic değil gerçekçi", ya da kullanıcı bir sahneyi sinema gibi değil

$ npx skills add ecinaro/brainlab-realistic-output
Skill anzeigen→

text-to-handwritten-notes

github.com

Turn any content (a PDF, pasted text, a solution, a web page, a photo of a page or whiteboard) into pages that look written by hand, either as a phone photo of a real notebook page (squared, lined or Seyès paper) or as a clean iPad page, exported as PDF or PNG. Writes in a bundled default handwriting, or in the user's own handwriting learned from photos of notes they already wrote. Maths is drawn by hand (fractions, sums, integrals, roots, intervals, Greek). Use when the user asks for handwritten notes, notes "à la main", notebook or iPad-style notes, or something written in their own handwriting.

$ npx skills add yahyaraissouni/text-to-handwritten-notes-skill
Skill anzeigen→

game-marketing-images

github.com

Produce store screenshots, promotional compositions, character cutouts, and animated headers from an existing game's assets and truthful gameplay captures. Use for a game press kit or store-image campaign; not for inventing a new game's visual identity.

$ npx skills add thomasbutyn/game-marketing-images-skill
Skill anzeigen→

comic-master-tooling

github.com

Comic-specific production tooling: using ComfyUI for consistent panels, vector lettering (balloons, tails, bursts, captions, SFX), detection on your own art, text on props like badges and signs, QA and page assembly. Use alongside comic-master.

$ npx skills add quantiqueue/comic-skills
Skill anzeigen→

brand-identity

github.com

Create a young company's brand identity from its positioning: a brand core, a name shortlist with the trademark and domain checks to run, a voice, a visual direction (colors with checked contrast, typefaces with open licenses) and simple SVG logo sketches to hand to a designer. Use when a founder needs a name, a logo, a brand or a visual identity. Triggers: "brand identity", "startup name", "logo", "branding", "color palette", "identité de marque", "nom de marque", "trouver un nom", "charte graphique", "identidad de marca", "nombre de marca", "logotipo", "paleta de colores".

$ npx skills add teodstr/garage
Skill anzeigen→

banner-compositor

github.com

Build wide header and banner images (subreddit, Discord, YouTube channel, forum, wiki, fan site, event or team page) by compositing real photos, stills, posters and logos into one strip with soft seams, with logos and title wordmarks kept exactly as published. Use this skill whenever the user wants a banner, header, cover image or collage strip made or updated from existing images, wants several subjects (films, games, teams, products, releases) combined into one banner, asks for "a prompt for an image AI" to make such a banner, or needs a logo cut out of a poster for one. Also use it to rebuild an earlier banner when new images arrive. Not for charts, UI mock-ups, or artwork drawn from scratch.

$ npx skills add sivarajd/banner-compositor
Skill anzeigen→

repoviz

github.com

Scan any codebase or logic and instantly generate rich visual diagrams—including architecture diagrams, flowcharts, Sankey diagrams, sequence diagrams, and mind maps—via zero-dependency cloud rendering URLs. Outputs ready-to-embed Markdown links by default, with on-demand export to local SVG, PNG, or standalone interactive HTML reports. Use when the user asks to visualize a repo, map architecture, trace data flows, draw a flowchart/Sankey/sequence diagram, or document system design.

$ npx skills add qiakai/repoviz
Skill anzeigen→

material-design

github.com

Apply the Material Design style (2014–present) when generating UI or graphic layouts. Load when the user asks for: material design, material ui, google material, elevation, floating action button, fab.

$ npx skills add musicalbee/design-ontology
Skill anzeigen→

brand-identity-design

github.com

End-to-end brand visual identity (VI) design methodology — a six-stage workflow from brand soul discovery to visual identity system building and multi-touchpoint brand manual delivery. Covers core concept, logo design, color strategy, typography, graphic language, and touchpoint application. Use when: designing brand identity, visual identity systems (VI), logo design, brand guidelines, or brand manuals. Keywords: brand identity, VI设计, 品牌视觉系统, 品牌VI, logo design, 品牌设计, 品牌手册, brand guidelines.

$ npx skills add not-narleeek/SKILLS
Skill anzeigen→

find-brand-logo

github.com

Find and retrieve existing company, product, developer-tool, self-hosted app, AI provider, and cryptocurrency logos with provenance across public catalogs. Use when a task needs real brand assets or a missing logo contributed; excludes designing or redrawing a logo.

$ npx skills add lxf82dev/open-brand-assets
Skill anzeigen→

elite-design-director

github.com

Unified design intelligence for websites, landing pages, sections, dashboards, components, redesigns, design systems, critiques and polish passes, plus a vector engine for logos, SVG illustrations, icon systems and raster-to-vector tracing, on any stack (WordPress/Elementor, WooCommerce, Shopify, React/Next.js, Tailwind, plain HTML/CSS). Acts as creative director + senior product designer + design engineer. Researches references (Dribbble, Pinterest, live sites), sets brand-specific art direction, builds token systems, prevents generic "AI-looking" output, keeps new work consistent with the existing site, and verifies with headless-browser screenshots, accessibility checks, SVG validation and a calibrated scorecard.

$ npx skills add khizerahmadchohan/elite-design-director
Skill anzeigen→

drawio-visuals-skill

github.com

Create and edit native Draw.io diagrams, explain processes through concrete editable objects and visible transformations, reconstruct diagrams from source material, and inspect, patch, version, or export .drawio files. Use for diagram images and HTML previews with official-renderer QA and layout/topology checks; native .drawio files are intermediate artifacts by default.

$ npx skills add henry867/drawio-visuals-skill
Skill anzeigen→

marca-da-empresa

github.com

Creates and applies the user's company brand identity (colors, fonts, logo, bilingual voice in English and Portuguese, words to use and avoid) in everything Claude produces, such as proposals, presentations, spreadsheets, documents, social posts, web pages and messages. When the user asks for the brand or brand guide, runs a short interview, pulls colors from the logo or Instagram screenshot, checks WCAG contrast and saves the brand manual in the company folder. When a manual exists, applies it automatically to every customer-facing deliverable. When no manual exists and the request is just a document or post, it does not interview, uses a neutral look and offers the manual in one line. Use when the user mentions brand, branding, brand guide, style guide, visual identity, logo, brand colors, tone of voice, or make it look like our company. Portuguese triggers - marca, identidade visual, logo, cores da empresa, tom de voz, manual da marca, deixar com a cara da empresa.

$ npx skills add ciceroaferreira/CRIE-AI-SKILLS
Skill anzeigen→

harness-brand

github.com

Use local brand packs to supply visual and writing guidance, templates and icons for branded slides, blogs and infographics. Build and validate packs, export brand contracts audit finished PowerPoint decks, and rebrand existing Google Slides or Docs with a selected local profile.

$ npx skills add dbbaskette/harness-brand-skill
Skill anzeigen→

anatomy-threejs

github.com

Build crafted, interactive isometric 3D technical figures in Three.js and GLSL that explain a technical concept, engine, algorithm, or system as a physical bench-top instrument, cutaway rig, or mechanical assembly. Features four-tone quantized cel shading with top-face bias, inverted-hull silhouettes, sharp crease lines, a 5-theme framed card with live telemetry readout, spring-eased kinematics, direct 3D part manipulation, bespoke physical controls, optional cutaway and exploded views, and an automated 3D spatial & visual verification pipeline. Use whenever someone wants an interactive 3D technical diagram, an isometric Three.js figure, a "Linear-style" or "Anatomy-style" 3D explainer, or a physical mechanism visualization with real-time WebGL shaders.

$ npx skills add Theblackcat98/anatomy-threejs
Skill anzeigen→

image-to-vector

github.com

Recreate a logo, icon, character, mascot or emblem from a reference image as accurate SVG vector art, restyle it (flat icon, soft 3D, cinematic metal, recoloured editions), and optionally animate it (idle, thinking, talking with lip-sync, paused). Traces the reference's own pixels instead of drawing by eye, proves the match with an overlay, and gets the design approved through previewed variants before building it into anything. Use this whenever someone wants an SVG, vector, icon or mascot that should look like a picture they have, converts a PNG or JPG to SVG, says a drawing "doesn't look like" the reference, asks to trace or outline an image, wants a recoloured or "dark edition" of a known design, or wants an animated mascot for a web page, an app or a Claude Code mod, even if they never say "trace" or "vector".

$ npx skills add SimsonCorreya/skills
Skill anzeigen→