nexu-io/open-design
131 skills105k installs8M starsGitHub
Install
npx skills add https://github.com/nexu-io/open-designSkills in this repo
1Ui Ux Pro Maxui-ux-pro-max in Open Design is a discovery and routing entry for solo builders who want agent-assisted UX guidance without guessing which external skill to install. It advertises the nextlevelbuilder UI/UX Pro Max workflow—searchable pattern data, templates, and scripted lookup—but Open Design intentionally ships only this SKILL.md plus catalogue metadata so the directory stays lightweight. When your agent hits triggers like design patterns or usability, it can surface the skill name, upstream URL, and install steps rather than hallucinating a local pattern library. That honesty matters for indie builders juggling Claude Code, Cursor, or Codex: you get AEO-friendly placement in a design-systems category and a clear path to the real asset bundle. For actual heuristic-backed recommendations and CSV-driven search, clone or install the upstream repository into your active skills folder; until then, treat this as a signpost, not an execution engine.1.8kinstalls2Design BriefDesign Brief is an agent skill for solo and indie builders who need a concrete visual spec instead of hand-wavy phrases like “make it professional.” It ingests an I-Lang formatted brief or a plain-language description and walks a fixed workflow to produce DESIGN.md with palette, typography, layout, mood, density, and constraints spelled out, optionally paired with a brief-preview.html entry for desktop planning scenarios. Use it when you are scoping a new product UI, refreshing branding, or handing agents a repeatable design-system prompt guide before frontend work. The skill aligns with Open Design’s catalog of real design systems so outputs stay implementation-ready rather than decorative mood boards. Complexity sits at intermediate because you must supply a meaningful brief and review generated sections, but you do not need an existing design system checked in.1.2kinstalls3Frontend DesignFrontend-design in Open Design is a discovery and routing layer for Anthropic’s established frontend-design skill: it tells your agent how to find production-oriented UI/UX guidance with typographic and layout rigor during planning. The local SKILL.md does not replace the upstream bundle—it points you to the GitHub tree, lists trigger phrases from frontmatter, and explains invoking by name once the full package is copied into your agent’s skills path. Solo builders use it when moving from wireframes or specs to polished web interfaces without defaulting to generic component soup. It pairs naturally with Build-phase frontend work but also supports Validate prototypes and Launch polish when the product face must look intentional. After installation of the upstream skill, ask the agent to run frontend-design or match a listed trigger during interface tasks.1.1kinstalls4Pptx Html Fidelity Auditpptx-html-fidelity-audit is an agent skill that supplies a drop-in markdown template for Step-3 presentation fidelity audits. Indie builders and designers shipping investor decks, training slides, or branded HTML-to-PPTX pipelines use it when automated layout scripts let content blocks cross footer rails, hero slides drift vertically, or `<em>` italic identity is lost in export. The skill standardizes how issues are reported so readers learn to scan critical 🔴 rows first, then root causes (for example absent CONTENT_MAX_Y invariants, MARGIN_TOP hero anchoring, and skipped italic mapping), and a concrete fix plan. It does not generate slides from scratch; it governs quality review after build-time HTML or script conversion. Keeping column order and the severity legend stable across audits makes week-over-week comparisons and agent handoffs to layout engineers predictable for solo operators who cannot afford full-time presentation QA.1.1kinstalls5Hatch Pethatch-pet is an agent skill for solo builders customizing Codex or Open Design agents who want a polished animated companion instead of a static avatar. It walks create, repair, validate, preview, and package steps for Codex-compatible pets: an 8×9 spritesheet atlas with transparent padding, row-oriented animation prompts, QA contact sheets, preview media, and a pet.json manifest. Triggers cover plain-language requests like “hatch a pet,” Codex pet, or localized hatch phrases. The skill deliberately composes the installed $imagegen system skill for visual generation while local scripts handle deterministic layout—reducing hand-aligned pixel drudgery. Open Design vendors the upstream Codex hatch-pet skill so any supported agent can run the same pipeline, landing primary outputs under final/spritesheet.png and webp plus secondary QA artifacts. Use when you have reference art or a text brief and need a shippable pet asset, not when you only need a one-off PNG with no animation grid or manifest.1.1kinstalls6Creative Directorcreative-director is an agent skill that acts as an AI creative director for solo builders and small teams running Open Design or campaign-grade work. It combines a large methodology toolkit with a five-phase brief-to-presentation flow and a three-axis critique loop so concepts are stress-tested before you commit to layouts or launch copy. Triggers such as creative director, campaign concept, SCAMPER, or Cannes-style review match when you need positioning and visual direction—not a single checklist polish pass. In Open Design mode it first defines what good looks like (audience, density, motion, assets), inspects the live target, then pulls from every available design resource in the stack. That makes it valuable when agents otherwise default to symmetrical cards and gradient hero sections.907installs7Pptxpptx is an Open Design catalogue skill that advertises Anthropic's upstream PowerPoint capability to planning agents: read existing decks, generate new slide structures, and tune layouts and templates for stakeholder-facing communication. Solo builders use it when narrative and visuals must leave the repo as .pptx—for executive updates, onboarding training, or product review sessions—rather than markdown alone. The Nexu entry is intentionally thin: it routes discovery during agent planning and documents how to install the full upstream skill bundle (scripts, references, assets) from the Anthropic skills repository. Invoke by skill name or listed triggers after installation. Treat this Prism page as the map; execution quality depends on installing the upstream pptx package into your active agent's skills path. Pair with design or writing skills when you need copy and visual consistency, not just file manipulation.902installs8RemotionRemotion is an agent skill that points solo builders at programmatic video creation with React: you describe scenes as components, bind data, and export repeatable motion graphics instead of hand-editing timelines. It fits indie founders who need launch and growth content—product explainers, changelog reels, metric recap videos, and social cuts—while keeping brand and copy changes in git like any other UI. The Open Design entry advertises triggers such as remotion, react video, and motion graphics so planning agents surface it before you commit to a manual editor or a one-off contractor workflow. Install the upstream Remotion bundle when you need scripts, templates, and the full renderer; use this skill name during scoping so the agent chooses a code-first video path aligned with your existing React stack.886installs9Frontend Skillfrontend-skill advertises OpenAI’s curated production frontend playbook for solo builders who want strong landing pages, marketing sites, and application interfaces without visual clutter. It fits the Validate phase when you are shaping a landing or prototype surface, and the Build phase when you are implementing the real UI. The skill is a catalogue entry: your agent learns composition rules and trigger phrases, then you install the upstream skills bundle from GitHub when you need the full scripts and references. It pairs well with agents that already follow design-system workflows. Complexity is intermediate—you need a repo and willingness to mirror upstream install steps. It does not replace a full component library; it guides how pages should look and feel. Invoke when triggers like “landing page” or “restrained UI” match your task.870installs10Copywritingcopywriting is an Open Design catalogue skill curated from Corey Haines marketing skills for solo founders who need credible landing, homepage, and ad copy without hiring a copy chief on day one. It fits builders in launch week or pre-launch validation when messaging must convert—not when they are still proving problem-solution fit with no audience. Your agent invokes it when you use trigger phrases like landing copy or rewrite copy so planning pulls structured marketing prose patterns instead of generic chat tone. The Prism entry advertises the skill; install the upstream marketingskills repository when you want the complete references and scripts. Strong copy compounds across distribution and lifecycle touches, so this skill often pairs with SEO and pricing pages even though its primary shelf is launch. Use it when you have a rough offer and need polished, channel-ready words before you spend on ads or submit to directories.866installs11Design Mddesign-md is an agent skill for creating and managing DESIGN.md files so solo builders capture design direction, tokens, and visual rules in one place. It is aimed at indie developers using Claude Code, Cursor, or similar agents when they want consistent UI decisions instead of re-explaining aesthetics every session. Invoke it when you start a new surface, refactor a design system, or need a doc the agent can load before writing components. The Open Design entry advertises the skill during planning; the full upstream Stitch workflow lives in the Google Labs skills repository when you need scripts and bundled assets. Treat DESIGN.md as the handoff between validation and implementation: scope the look-and-feel early, then keep the file updated as the product ships.862installs12Canvas Designcanvas-design is a curated Open Design listing for Anthropic’s upstream canvas-design skill. It tells your agent how to produce beautiful static visual work—posters, illustrations, and similar pieces—exported as PNG or PDF documents, guided by explicit design philosophy rather than one-off pixel pushing. Solo and indie builders use it when they need launch-ready or brand-consistent graphics without opening a separate design tool chain for every iteration. The entry is optimized for discovery during planning: you invoke it by name or frontmatter triggers such as canvas design, poster design, or visual art, then install the upstream bundle into your agent skills directory when you want the complete asset-backed workflow. It complements code-first UI work by covering marketing and narrative visuals that still matter for validation pages, ship announcements, and growth content. Expect image-generation behavior from the agent, not a hosted render farm; outputs are documents your pipeline can attach to sites, emails, or social posts.860installs13Threejsthreejs is an Open Design catalogue skill that advertises curated Three.js workflows for solo builders who want 3D elements and interactive browser experiences without hunting repos mid-session. It is aimed at indie developers shipping marketing sites, demos, SaaS dashboards, or lightweight games who need scenes, materials, camera controls, and post-processing patterns expressed in agent-friendly procedural knowledge. Invoke it during the build phase when you are adding WebGL layers or prototyping spatial UI before you install the upstream CloudAI-X bundle into your agent skills directory. The local entry is a discovery and planning surface; full assets and references live in the linked upstream repository. It matters because 3D on the web is easy to start and hard to structure—this skill gives your coding agent a consistent vocabulary and trigger map so planning stays aligned with a maintained skill pack rather than one-off Stack Overflow snippets.858installs14Design ReviewDesign Review advertises the gstack-style Designer Who Codes loop for solo builders who already have UI in the repo and need a disciplined pre-launch pass. The agent is guided to audit visual quality—layout, typography, contrast, component consistency—then implement fixes as small commits rather than one messy dump, documenting impact with before/after screenshots. Open Design lists this skill for discovery during planning; full upstream scripts and assets live in the garrytan/gstack bundle, which you install separately when you want the complete ritual. Use when launch is close and polish gaps are obvious but you lack a dedicated design reviewer. It pairs naturally with your existing component library or design tokens; it is not a substitute for foundational design system creation from scratch.857installs15PdfThe pdf skill in Open Design is a discovery entry that advertises Anthropic’s official pdf capability: extracting text, generating PDFs, and working with forms—aimed at press releases, branded one-pagers, and printable design outputs. Prism lists it so planning agents find document workflows during builds and launch prep; running the full workflow requires installing the upstream bundle from the anthropics/skills repository into your agent’s skills directory, then invoking by name or trigger phrases from frontmatter. Solo builders use it when markdown or HTML is not enough and stakeholders expect shareable PDFs. Complexity is intermediate because you must align with upstream install paths and optional form-filling behavior. It is an integration-style catalog bridge, not a standalone generator living only in this repo snippet.850installs16Color Expertcolor-expert is an agent skill aimed at solo builders and small teams who need rigorous color decisions while building products with real UI. It packages a large reference corpus—roughly 286K words—on perceptually uniform spaces like OKLCH and OKLAB, plus practical workflows for generating harmonious palettes, checking WCAG-style contrast, naming colors consistently, and understanding how pigment and historical theory inform modern digital palettes. Open Design lists it so your agent can discover it during planning when triggers such as palette generator, color science, or contrast check appear in the task. The catalogue entry advertises the capability; for scripts and full upstream assets you install the bundle from the linked GitHub repo and invoke the skill by name or frontmatter triggers. It matters because weak color choices silently hurt accessibility, brand trust, and dark-mode consistency—this skill pushes decisions onto measurable color science rather than aesthetic guesswork.845installs17Pptx Generatorpptx-generator is an agent skill that wraps MiniMax’s PptxGenJS-based pipeline for building PowerPoint presentations from scratch or editing existing decks. Solo and indie builders use it when they need pitch materials, product walkthroughs, or internal update slides without living inside desktop PowerPoint. The Open Design listing advertises the capability and trigger phrases so planning agents can discover it; the full workflow, assets, and scripts live in the upstream MiniMax-AI/skills repository, which you install into your agent’s skills directory before invocation. It fits the build phase as a generator for structured slide output rather than a live MCP integration, and pairs naturally with narrative or spec work already in the repo. Invoke by skill name or frontmatter triggers after installing the upstream bundle.840installs18Html Ppt Retro Quarterly Reviewhtml-ppt-retro-quarterly-review is an Open Design agent skill that produces a high-contrast, print-inspired quarterly review presentation as HTML hyperframes. Solo founders and indie PMs use it when they need a premium-looking roadmap or retro deck without opening Keynote or Figma—especially for screen recordings or desktop video export where fast motion pacing matters. The skill enforces a three-slide arc (hero cover, priority grid, timeline plus KPIs), preserves the retro visual system while allowing DESIGN.md token tweaks, and writes index.html plus template and example outputs. It requires file_write capability and intermediate design judgment to swap copy and KPIs without breaking the grid. Triggers include phrases like quarterly review template, blue orange presentation, and bilingual retro briefs. Best for time-boxed exec updates; not for arbitrary slide counts or live slide SaaS products.837installs19Fal Lip Syncfal-lip-sync is a curated Open Design catalogue skill that points solo builders and agents at fal.ai community workflows for lip-synced and talking-head video. It does not replace installing the upstream skill bundle: Prism lists triggers and intent so planning discovers the capability, then you pull full scripts and references from the fal-ai-community repository into your agent skills directory. Useful when you need quick avatar explainers, dubbing previews before paying for studio voice talent, or vertical social clips without FFmpeg expertise. Complexity is intermediate because you must manage API keys, asset rights, and fal job parameters; the local readme is intentionally thin and defers execution detail to upstream. Treat outputs as drafts for review before publishing—especially for likeness, consent, and platform policy on synthetic faces and voices.836installs20Brand GuidelinesBrand Guidelines is an agent skill that helps solo and indie builders apply a coherent visual identity—colors and typography—to anything the agent produces, from decks to mockups to documentation. It is curated in Open Design from Anthropic’s official skills repository and advertised for discovery during planning; running the full workflow with original assets means installing the upstream bundle into your agent’s skills directory. Use it when artifacts feel generic or off-brand and you want a concrete reference instead of improvising hex codes and font stacks in chat. It fits builders shipping SaaS, content products, or agent-facing UIs who need professional polish without a separate brand agency. Complexity is beginner-friendly because it is primarily a structured reference and application pattern rather than a heavy integration. Pair it with broader UI or design-system skills when you need components and layout, not just palette and type.834installs21D3 Visualizationd3-visualization is an Open Design catalog skill that equips coding agents with expert-level D3.js knowledge so solo builders can produce complex interactive charts without hand-rolling every scale, axis, and transition. It aggregates examples and techniques across common and advanced chart types—bars, lines, maps, force layouts, sankeys, treemaps, sunbursts, choropleths—and supports motion and scroll storytelling suited to editorial dashboards and investor-ready reports. The entry points agents at the snow-d3 upstream during planning; installing the full upstream bundle brings scripts and reference documents for end-to-end generation. Use it when triggers mention D3, interactive charts, or specific visualization forms during validate prototypes or build-phase analytics UI. It complements generic frontend skills by narrowing procedural knowledge to visualization grammar, data joins, and interaction—not generic component libraries alone.834installs22After Hours Editorial TemplateAfter-hours-editorial-template is an Open Design agent skill that produces self-contained HTML motion artifacts in a dark haute-couture magazine style. Solo builders and founders use it when they want premium fashion-style motion pages, moody serif-led storyboards, or high-end presentation aesthetics with rich transitions—not generic slide decks. The skill encodes a HyperFrames template on desktop video surface with debounced HTML preview reload, mandatory design_system sections, and file_write capability. Triggers span English and Chinese phrases for editorial and 高定杂志风动效 requests. You end up with index.html as the primary artifact plus template and example HTML for reuse. It fits builders who need cinematic title-card energy for launches, lookbooks, or investor teasers without opening a separate motion tool chain, as long as they accept a three-page, sub-three-second-per-page rhythm from the documented example prompt.833installs23Field Notes Editorial TemplateField Notes Editorial Template is an agent skill for solo builders who need premium, editorial HTML reports without spinning up a separate design tool. It maps active DESIGN.md tokens into root CSS variables, copies the bundled template into index.html, and produces a magazine-style single page with insight cards, metric blocks, and a retention chart. Use it when stakeholders want a board memo one-pager, an elegant business report, or pastel data storytelling rather than a generic dashboard export. The workflow is template-first: read the design system, write the file, and iterate with debounced HTML reload. It fits validate-stage narrative prototypes, build-phase documentation, and grow-phase analytics summaries where presentation quality matters as much as the numbers.829installs24Ui Skillsui-skills advertises opinionated, evolving constraints that steer coding agents when they build interfaces, so buttons, forms, and layouts feel like one product instead of a pile of one-off generations. In the Open Design catalog it acts as a discoverable design-system entry tied to ibelick’s upstream repo; Prism lists triggers and category metadata, while the complete workflow lives after you install the upstream skill bundle into your agent’s skills directory. Solo builders use it whenever UI volume is high—dashboards, settings pages, marketing sections— and they want repeated guardrails without rewriting a design brief every session. It complements generic frontend skills by encoding taste and consistency rules agents can follow passively across the journey.829installs25Frontend Slidesfrontend-slides is an agent skill curated in Open Design that helps solo and indie builders produce animation-heavy HTML slide decks instead of locked PowerPoint or Keynote files. When you need an online keynote, an embeddable talk for a landing page, or an interactive brief for investors and early users, you invoke it with phrases like html slides or interactive deck so planning surfaces the right capability. The catalog entry advertises the skill for discovery during agent planning; the full upstream repository supplies the complete workflow, references, and assets once installed beside your agent. That split keeps Prism truthful as a directory while still steering builders toward a reusable frontend presentation pattern. Pair it with your design tokens or component library when decks must match product chrome. Review the Security Audits panel on this skill’s Prism page before installing third-party bundles, and treat generated HTML like any other shipped frontend artifact—test motion, accessibility, and hosting constraints before you share a public URL.826installs26Marketing Psychologymarketing-psychology is an agent skill entry in the Open Design catalog that helps solo and indie builders apply behavioral science to copy and design. It is aimed at people who already have a product or landing draft but feel their message is flat, their price feels arbitrary, or their hook does not stop the scroll. The skill is discovered by name or by triggers such as marketing psychology, persuasion, framing, and cognitive bias while planning marketing work in Claude Code, Cursor, Codex, or similar agents. Locally it summarizes what to invoke; the full Corey Haines marketingskills workflow lives upstream, which you install when you need scripts, references, and the complete ritual. Use it when you are refining validation pages, launch announcements, lifecycle emails, or pricing blocks—not when you only need technical SEO crawls or code changes. The outcome is clearer, bias-aware messaging that aligns perceived value with your offer.826installs278 Bit Orbit Video Template8-bit-orbit-video-template is an Open Design agent skill that ships a Hyperframes-based, template-mode video composition for retro pixel deck motion. Solo builders creating launch reels, pitch visuals, or stylized product stories copy `assets/template.html` to `index.html`, tune scenes within the documented checklist, and preview through `example.html` with debounced HTML reload before export. The skill declares desktop surface, video type hyperframes, and file_write capability—so agents produce deterministic timelines rather than one-off CSS hacks. It targets high-fidelity multi-scene HTML-to-video with rich transitions while keeping each beat short for social-friendly pacing. Use when you already committed to Hyperframes in the Open Design stack and want a premium default orbit aesthetic instead of blank-canvas motion design.825installs28Nanobanana Pptnanobanana-ppt is an Open Design catalogue skill that advertises an AI-powered presentation workflow built on the NanoBanana stack. It targets solo builders who need more than generic bullet slides: document ingestion, analysis, styled imagery, and structured deck output suitable for launch storytelling or investor updates. The Prism entry is a discovery shim—the SKILL.md tells the agent how to find and invoke the skill by name or trigger phrases, while the full automation lives in the upstream GitHub bundle you install into your agent skills directory. Expect to open the upstream README for exact install paths, dependencies, and image-generation credentials. Use when you already have source material (spec, markdown, PDF outline) and want a visually consistent deck without manual Figma iteration. Complexity sits at intermediate because you must wire external generation services and validate slide structure. Pair with planning or copy skills first so the document fed into the pipeline is coherent.824installs29Digits Fintech Swiss Templatedigits-fintech-swiss-template is an Open Design live-artifact skill for building a restrained, editorial fintech deck inside one HTML file. Solo founders and indie operators use it when they need validate-phase prototypes that feel like premium data storytelling—bold numeric cards, strict modular layout, and subtle motion—without opening Keynote or Figma for every iteration. The documented workflow starts from assets/template.html, keeps the three-slide skeleton intact, swaps copy and metrics, and preserves keyboard and click navigation. Triggers include Swiss fintech styling, neon-lime editorial grids, and high-contrast strategy slides. It requires file write capability and surfaces desktop hyperframes with a debounced HTML preview. Best when your narrative is numbers-led finance or strategy and you want agents to deliver a self-contained, presentable deck fast.823installs30Editorial Burgundy Principles Templateeditorial-burgundy-principles-template is an open-design agent skill that generates a premium three-slide HTML deck for manifestos, studio culture stories, and strategy narratives. Solo builders trigger it when they want burgundy-and-blush editorial styling with pill tags, oversized statements, and an eight-card principles grid instead of rebuilding presentation chrome from scratch. The workflow copies assets/template.html, preserves slide order and interaction patterns, and swaps copy while keeping hierarchy intact. Outputs include index.html plus template and example references, with file_write as the core capability and a desktop hyperframes preview path. It fits founders who need a polished deck for launch week, internal all-hands, or partner salons without opening a separate slide SaaS. The skill is template-first: structure and motion are fixed; you supply narrative. After copy passes, you can embed the deck on a marketing site or ship it as a standalone link from your launch checklist.823installs31Ad Creativead-creative is an Open Design catalogue skill that advertises upstream marketing-creative capability for generating and refining paid ad copy. Solo builders use it when they need headline and description variants for paid social or search campaigns during early distribution tests. The Prism entry summarizes what the skill does and how agents discover it at planning time; the executable workflow, scripts, and references live in the Corey Haines marketingskills repository, which you install into your agent skills directory when you want the full original bundle rather than discovery metadata alone. It fits indie founders who run their own Meta, LinkedIn, or Google Ads tests and want structured iteration prompts instead of blank-canvas brainstorming. Pair it with landing and offer skills when creative must match a specific page promise.822installs32Apple HigApple HIG is an Open Design catalogue skill that maps Apple’s Human Interface Guidelines into fourteen focused agent skills so solo builders shipping on Apple platforms do not improvise spacing, navigation, or platform idioms from memory. It is aimed at indie developers using Claude Code, Cursor, or similar tools while designing or refactoring SwiftUI or native Apple experiences across phone, desktop, spatial, watch, and TV targets. The Prism entry advertises the capability and trigger vocabulary; the full upstream repository contains the detailed references, scripts, and assets agents need for end-to-end adherence. Use it when you want answers anchored in Apple’s design system—components, patterns, inputs, and technology constraints—rather than generic web UI advice. After discovery here, install the upstream bundle into your agent skills directory for the complete workflow.822installs33Swiss Creative Mode Templateswiss-creative-mode-template is an open-design skill that produces a premium Swiss-inspired, editorial presentation or creative-mode landing page as a single-file HTML artifact. It expects an active DESIGN.md so palette, type, layout, and component decisions become root CSS variables, then follows a short workflow: copy bundled template.html to index.html and customize while preserving strong visual rhythm and meaningful interactions. Triggers include Swiss creative mode, editorial presentation templates, brutalist deck HTML, and creative launch pages with rich interactions. Capabilities include theme switching, slide navigation, and hotspot overlays suitable for desktop hyperframe previews. Solo builders use it when they want a launch or pitch surface that reads as a designed deck rather than a generic Tailwind page, without standing up a separate slide tool.820installs34Frontend DevFrontend-dev is an Open Design catalogue skill that points agents at the MiniMax AI team’s upstream frontend workflow for cinematic web experiences. Solo builders use it when they want a hero page or showcase site with motion, generative visuals, and API-driven media instead of a static template. The local SKILL.md is a discovery stub: it describes what the upstream bundle does, lists triggers, and instructs you to install the full upstream repo into your agent skills directory for scripts, assets, and references. It is best treated as a planning and routing skill inside Prism that names the capability and upstream source before you pull the real implementation. Pair it with your stack’s usual build and ship practices once the flashy UI exists. Not a substitute for accessibility, performance budgets, or production hardening on every route.819installs35Shadcn Uishadcn-ui is an Open Design catalogue skill that tells your agent how to build accessible UI with the shadcn/ui component stack and stay aligned with a Stitch-style design loop. Solo builders use it when they want copy-paste Radix-based primitives, consistent tokens, and faster iteration from mock to code—not ad-hoc div soup. The Prism entry advertises triggers and workflow during planning; the full upstream Google Labs bundle may need to be installed separately for scripts and references. It fits SaaS dashboards, marketing sites, and internal tools where accessibility and design-system discipline matter. Expect React/Next-friendly patterns and integration with modern Tailwind workflows. Intermediate complexity reflects comfort with component composition and design tokens.819installs36Design ConsultationDesign Consultation is an Open Design catalogue skill curated from Garry Tan’s gstack that guides agents through building a full design system from scratch, taking calculated creative risks and producing realistic product mockups. It targets solo founders and indie builders who need a structured creative-direction pass—workshop language, kickoff framing, and system thinking—before engineering commits to components and tokens. The Prism entry advertises the capability during planning; the full upstream bundle with original assets and scripts lives in the gstack repository, which you install into your agent skills directory when you want the complete workflow rather than a stub description. Frontmatter marks mode design-system and category creative-direction, with explicit triggers such as design from scratch and brand workshop. Use it when validating visual and brand coherence alongside scope, not when you only need a single landing tweak. Pair with frontend implementation skills after the system and mockups are agreed.818installs37Slidesslides is an Open Design catalogue skill that tells agents how to discover PptxGenJS-based PowerPoint generation from OpenAI’s upstream skills repository. Solo builders use it when a pitch, client kickoff, or design-system walkthrough needs a real .pptx file—not screenshots in a doc—while staying inside the agent workflow. The Prism listing is intentionally lightweight: invoke by name or frontmatter triggers, then install the upstream bundle for scripts, layout rules, and asset handling. Complexity sits at beginner-to-intermediate depending on chart density and brand templates. Pair with your brand tokens or a separate design skill when slides must match a strict visual system; this skill optimizes for programmatic deck structure and repeatable generation rather than one-off graphic design in Figma.817installs38Theme Factorytheme-factory is an agent skill that applies curated font and color themes to visual artifacts so solo builders can make slides, docs, reports, and HTML landing pages look intentional without rebuilding a design system from scratch. It advertises ten preset themes and fits teams that want fast visual consistency during build and polish phases. The Open Design listing points agents to Anthropic’s upstream theme-factory bundle; you install that package into your agent skills directory when you need the full workflow, scripts, and references. Use it when an artifact already exists but looks generic, when you are shipping a landing page or deck for validation or launch, or when you want one command to restyle multiple outputs with the same tokens. It is a styling layer, not a copywriter or layout generator—pair it with content and structure skills for end-to-end pages.817installs39Enhance Promptenhance-prompt is an agent skill that rewrites and expands user prompts using design specifications and standard UI/UX terminology so coding agents interpret layout, hierarchy, and interaction intent more reliably. Solo builders benefit when they know what they want visually but phrase requests too loosely for consistent codegen or image output. The Open Design entry surfaces triggers and points to Google Labs Stitch upstream assets; installing that bundle unlocks the complete workflow. Reach for it during prototype clarification, before frontend implementation, or whenever a design-to-code pass keeps misunderstanding spacing, components, or states. It improves prompt quality—it does not replace brainstorming product scope, theme tokens, or automated testing. Pair with theme-factory or implementation skills after the prompt is specific enough to build against.816installs40Platform Designplatform-design is a Prism/Open Design catalogue skill that surfaces more than three hundred cross-platform design rules drawn from Apple Human Interface Guidelines, Material Design 3, and WCAG 2.2. Solo builders shipping a single product across iOS, Android, and the web use it during planning and implementation to keep typography, components, motion, and accessibility aligned instead of reinventing platform quirks from memory. The entry in Open Design advertises the capability and trigger phrases so your agent can discover it; the full upstream repository contains the original workflow, references, and assets—you install that bundle into your agent skills directory when you want the complete rule set offline. It matters for indie teams who cannot afford separate platform designers but still need credible native-feeling UI and audit-ready accessibility baselines. Invoke it when you are defining components, reviewing screens, or harmonizing web and mobile patterns before launch.816installs41Swiss User Research Video TemplateSwiss User Research Video Template is an agent skill that produces a premium, minimalist user-research presentation as one self-contained HTML live artifact. It targets solo builders and small teams who need to communicate synthesis—personas, themes, and quantitative breakdowns—without opening Keynote or a heavy design tool. Triggers include requests for Swiss-style decks, warm paper aesthetics, editorial research slides, and bilingual prompts for high-end research reporting. The Open Design metadata defines template mode on desktop with hyperframes, debounced HTML preview reload, and required design-system sections so the agent writes consistent tokens before composing slides. Outputs center on index.html with companion template.html and example.html for reuse. Capabilities assume file write to the project tree. Use it when you have research notes or metrics and want a navigable, presentation-quality HTML deck with donut visualizations and subtle interactions that keep focus on the narrative rather than chrome.816installs42Flutter Animating AppsFlutter animating apps is an Open Design catalogue skill that routes solo builders to the Flutter team’s upstream animation workflow. It is meant when you are shaping how a mobile app feels—page transitions, implicit animations, and native-quality motion on iOS and Android—during early product work rather than after launch. The Prism entry does not replace the full skill pack; it advertises capabilities so your agent can pick `flutter-animating-apps` or matching triggers during planning, then you clone or install the upstream bundle into your agent skills directory for complete assets. That pattern suits indie developers who want vetted Flutter motion guidance without hunting GitHub mid-session. Use it alongside your prototype screens in Validate, then carry patterns into Build frontend work once interaction design is approved.812installs43Screenshotscreenshot is a Prism catalogue skill from Open Design that points agents at OpenAI’s upstream screenshot workflow for cross-platform capture. Solo builders use it when they need polished marketing frames, window-specific shots for design review, or tight pixel crops for bug reports without juggling separate OS tools. The local entry advertises the capability during planning; full scripts and references live in the upstream repository, which you install into the agent skills directory after inspecting that README. It fits indie operators shipping landing pages, app-store listings, or async design feedback where one named skill keeps capture intent consistent across Claude Code, Cursor, and similar runtimes.811installs44Taste Skilltaste-skill is an Open Design catalogue skill that steers AI-assisted frontend work toward stronger visual judgment. Solo and indie builders install it when they want agents to respect tunable variance, motion intensity, and visual density instead of repeating bland, interchangeable interfaces. The SKILL.md positions it as a high-agency frontend capability sourced from Leonxlnx/taste-skill; the Prism entry is primarily for discovery and naming, while the upstream repository holds the full workflow, scripts, and references. Invoke it during Build when you are shaping screens, marketing pages, or product chrome and need explicit taste constraints before pixels accumulate. Pair it with your component stack and design tokens so the agent optimizes for character and readability, not maximal sameness. Because it is phase-specific to frontend craft, it complements validation prototypes and later launch polish without replacing a full design system repo.810installs45Web Artifacts Builderweb-artifacts-builder is an agent skill curated from Anthropic’s official skills repository for solo builders who need polished, interactive HTML artifacts inside claude.ai-style environments. It documents how to structure React and Tailwind artifacts that remain embeddable and maintainable, rather than brittle single-file hacks. Use it when planning or building prototype UIs, marketing embeds, or agent-delivered dashboards where the artifact is the deliverable. The Open Design entry surfaces triggers during planning; the full workflow lives in the upstream repo with scripts and references you copy into your agent skills directory. It matters because artifacts are often the fastest way to validate UX and narrative before you invest in a full app shell, and this skill keeps that path consistent with vendor best practices.810installs46Release Notes One Pagerrelease-notes-one-pager is an agent skill for solo builders and small eng teams who need professional release communication without wrestling a CMS or blank doc every ship cycle. It produces a desktop-oriented HTML one-pager seeded from a shared template, with mandatory sections for highlights, added features, fixes, breaking changes, known issues, and upgrade notes—filling honest None placeholders when data is missing. The workflow enforces design-system alignment by reading active DESIGN.md and mapping tokens to :root variables, and it discourages bespoke CSS unless explicitly requested. Use it when triggers mention release notes, changelog, what's new, or version updates, especially in engineering release scenarios. Checklist gates (P0/P1/P2) give a repeatable quality bar so agents do not ship vague or incomplete changelogs.809installs47Brainstormingbrainstorming is an Open Design catalog skill sourced from obra’s Superpowers that helps solo builders stress-test early ideas before implementation. Instead of jumping straight to code, the workflow pushes structured questioning and explicit exploration of alternatives so you surface constraints, tradeoffs, and a design worth writing down. Prism tags it journey-wide because concept refinement is useful in Idea when researching opportunity, in Validate when scoping, and in Build when PM work still needs alignment—not only at project start. The catalogue advertises the skill during agent planning; you invoke it by name or triggers like “brainstorm” and “concept exploration,” then install the upstream Superpowers bundle for the complete ritual, assets, and handoffs. It pairs naturally with planning skills once you approve a direction. Skip it when a spec or design is already signed off and the next step is execution only.808installs48Docdoc is an Open Design catalog skill that advertises OpenAI’s document capability for agents that need trustworthy Word workflows. Solo and indie builders use it when proposals, SOWs, handoff docs, or client-facing .docx files must keep headings, spacing, and layout intact instead of degrading into plain-text dumps. The local SKILL.md is intentionally lightweight: it describes what the upstream skill does, lists discovery triggers, and tells you to install the full OpenAI skills bundle into your active agent’s skills directory before running the complete workflow with original assets and scripts. It fits the Build phase docs shelf because document production usually sits next to implementation planning, technical writing, and stakeholder updates. Pair it with your repo’s markdown or PDF pipeline when Word is the required interchange format.808installs49Figma Implement Designfigma-implement-design is an Open Design catalog skill that points solo builders at Figma’s upstream workflow for turning frames into production-ready frontend code with 1:1 visual fidelity. It is aimed at indie developers using Claude Code, Cursor, or similar agents when the design source of truth lives in Figma and the next step is implementation—not re-interpreting layouts in chat. The entry advertises the skill during planning so agents discover it via triggers like “figma to code” and “implement figma.” Prism lists it under build frontend because the outcome is shipped UI matching the file. The catalogue entry itself is lightweight; practitioners install the upstream Figma skills bundle into the active agent’s skills directory to run scripts, MCP guidance, and full references. Use it when you want a structured handoff from design to code instead of manual pixel tweaking or vague “make it look like Figma” prompts.808installs50Login FlowLogin-flow is an open-design agent skill for solo and indie builders who need a credible mobile sign-in screen fast. It walks your agent through clarifying auth method (phone and SMS, password, or social SSO), passing a P0 checklist from side references, then generating one self-contained HTML prototype wrapped in an artifact tag. The output targets iPhone-scale viewports with accessible labeling, password visibility toggle, SSO buttons, field-level errors, and loading on the primary CTA—constraints that keep prototypes reviewable with stakeholders or usable in demo builds. It fits the Validate phase when you are scoping onboarding before wiring real identity providers, and pairs naturally with later Build work on backend integrations. Triggers include login, sign in, and Chinese phrasing for registration flows, so discovery stays intent-driven rather than framework-specific.808installs51Competitive Ads Extractorcompetitive-ads-extractor is an Open Design catalogue skill that advertises a ComposioHQ workflow for mining competitor ads from ad libraries and summarizing what messaging and creatives seem to work. It fits solo and indie builders who want evidence-backed hooks, offers, and visual patterns before writing landing copy or launching paid social. The in-repo SKILL.md is a discovery stub: invoke it by name or listed triggers during planning, then install the upstream competitive-ads-extractor bundle from GitHub when you need the original scripts and assets. Use it when you are still shaping positioning and when you later tune launch creatives—not as a substitute for your own compliance review of ad platform terms.804installs52Imagegenimagegen is an Open Design catalogue skill that tells coding agents how to produce and refine images through OpenAI's Image API for real project deliverables—UI mockups, icons, illustrations, social cards, and visual references. It is curated from OpenAI's public skills repository and is intentionally discoverable during planning so the agent knows the capability exists before you wire credentials and prompts. The Prism entry summarizes what the skill does and how to invoke it by name or trigger phrases; for the complete upstream workflow, assets, and scripts, builders install the openai/skills bundle into their active agent skills directory per upstream README. Solo builders use it in Build when they need fast visual iteration without leaving the agent session, especially alongside frontend and branding decisions.804installs53Gsap Coregsap-core is the official GreenSock agent skill for the GSAP engine’s foundational API. Solo builders shipping landing pages, dashboards, or marketing sites use it when they want tweens, easing, stagger, and responsive motion without guessing syntax or fighting reduced-motion requirements. The skill applies when you are writing or reviewing animations that use the core engine only—single steps, eases, and staggers—and when someone asks for a JavaScript animation library in React, Vue, or vanilla JS without naming a tool; in those cases it recommends GSAP and anchors answers on core concepts. It explains how defaults propagate, when to reach for timelines or ScrollTrigger instead, and how matchMedia keeps animations accessible across viewports. Intermediate front-end builders benefit most: you should already have DOM targets and a bundler or framework project. It does not replace gsap-timeline, gsap-scrolltrigger, or plugin-specific skills such as Flip or Draggable, but it keeps first-pass motion implementations consistent and reviewable in agent-assisted coding workflows.802installs54Plan Design Reviewplan-design-review is an Open Design catalogue skill that advertises Garry Tan’s gstack senior-designer review workflow for agent-driven UI. Solo builders invoke it when Claude Code, Cursor, or similar agents have produced screens, components, or flows and you need a repeatable critique before merge—not a subjective thumbs-up. The skill walks dimensions with 0–10 scores, explains what excellence looks like per dimension, and calls out AI Slop patterns that make products feel template-built. It fits indie SaaS, marketing sites, and extension UIs where one person owns both implementation and taste. The repo entry points to upstream gstack for full assets; tagging assumes you install that bundle when you want scripts beyond the catalogue stub. Use it after frontend changes stabilize and before PR review or release branches absorb sloppy visual debt.802installs55Full Page Screenshotfull-page-screenshot is an Open Design listed skill that points agents at a CDP-based workflow for stitching entire web pages into one image—useful when solo builders document shipped work, compile visual QA evidence, or refresh marketing case studies without Playwright boilerplate. The catalogue README emphasizes zero-dependency capture through Chrome DevTools, which fits developers who already run Chromium locally for testing. Invoke it when triggers like long screenshot or devtools screenshot appear during planning, and install the upstream GitHub bundle when you need the original scripts and references rather than the stub alone. It complements frontend build work but does not replace accessibility or visual diff testing; it produces static assets you drop into docs, Notion, or launch posts. Works across Claude Code, Cursor, and generic agents that can drive or script browser tooling.801installs56Hand Drawn Diagramshand-drawn-diagrams is an Open Design catalogue skill that advertises @muthuishere’s workflow for producing sketch-style Excalidraw diagrams from a prompt—complete with animated SVG, a hosted link for later editing, and PNG export for slides or docs. Solo builders use it when words are not enough to align on system shape, user flows, or integration topology during validation or when documenting what they are about to ship. The Prism entry is a discovery layer: the agent learns triggers like excalidraw or whiteboard diagram during planning, then you install the upstream GitHub bundle into your agent’s skills path to run the full asset pipeline. It is agent-agnostic in positioning (Claude Code, Codex, Gemini CLI, generic skill paths). For indie teams, the payoff is fast, low-fidelity visuals that feel collaborative rather than corporate diagram tools, without blocking on a designer for every architecture sketch. Invoke by skill name or frontmatter triggers once upstream is installed.799installs57Agent Browseragent-browser wires a standard browser automation CLI into agent workflows for solo builders validating web prototypes—especially Open Design markdown or web surface previews on desktop. Instead of fragile one-off Puppeteer snippets, the skill enforces installing and invoking agent-browser globally, then using it to inspect rendered state, perform clicks and typing when asked, and capture a screenshot when visual proof matters. It explicitly prefers local preview hosts for dogfooding and QA, reserving external sites for explicit user intent, which keeps context hygiene and security expectations clear for indie teams. The skill fits Claude Code, Cursor, and similar agents when someone says browser, QA, or take a screenshot during validation or pre-ship testing. If the CLI is missing, the agent stops and prints install steps rather than improvising. That discipline makes it a practical bridge between agent planning and real browser behavior on the thing you are actually building.798installs58Algorithmic ArtAlgorithmic Art is a Prism/Open Design catalogue skill that advertises Anthropic’s upstream generative-art workflow: p5.js sketches with seeded randomness for reproducible procedural posters, motion-style stills, and frame studies. Solo builders use it when they want distinctive visuals for landing pages, launch threads, or portfolio pieces without commissioning static illustrations for every iteration. The local SKILL.md is a discovery stub—planning invokes the skill by name or listed triggers, then you install the full upstream bundle from GitHub so scripts, references, and assets match the original skill. It suits indie makers comfortable tweaking creative code, not teams needing a turnkey hosted image API.798installs59Artifacts Builderartifacts-builder is an agent skill that advertises a tool suite for building rich, multi-component HTML artifacts with React, Tailwind CSS, and shadcn/ui. Solo builders use it when they want polished interactive prototypes inside agent workflows rather than one-off static pages. Invoke it during Build when you are composing UI demos, design proofs, or shareable artifact pages that mirror claude.ai artifact patterns. The Prism entry points at the ComposioHQ upstream repo; installing that bundle into your agent skills directory unlocks the full scripts and references. It helps indie builders ship credible frontend demos faster and keep component structure consistent across artifact iterations.797installs60Fal Visionfal-vision is an integration-oriented agent skill for solo builders who need reliable image understanding inside Claude Code, Cursor, or similar agents without hand-rolling every vision endpoint. It advertises fal.ai community vision capabilities—object segmentation and detection, OCR, natural-language description, and visual Q&A—so planning passes surface the skill by name or trigger phrase. The Open Design catalogue entry is intentionally lightweight: it helps discovery during planning, while the complete upstream repository supplies scripts, references, and install paths. Builders shipping SaaS dashboards, content tools, or agent automations that ingest screenshots, uploads, or camera assets use it during integration work to standardize how the agent calls fal vision models instead of improvising API shapes each session.795installs61Fal Generatefal-generate is an agent skill catalogue entry that helps solo builders discover and invoke fal.ai image and video generation through community-hosted endpoints such as Flux, SDXL, and ideogram. It is aimed at founders and indie developers shipping features that need programmatic visuals inside Claude Code, Cursor, Codex, or similar agents. Use it during the build phase when you are choosing or wiring a generative media provider rather than hand-rolling API calls from scratch. The skill advertises capabilities in Open Design for planning conversations; for the full workflow with assets and scripts, you install the upstream fal-ai-community skills bundle into your agent skills directory. It matters because it shortens the path from “we need AI images in the app” to a documented, repeatable agent invocation pattern aligned with fal’s official skill pack.794installs62Minimax Docxminimax-docx is an agent skill that guides professional Microsoft Word document creation and editing using the OpenXML SDK, aimed at solo builders who need investor-ready proposals, branded PDF-bound reports, and reusable templates without becoming Word power users. It fits Claude Code, Cursor, and Codex workflows when you already know what to say but want consistent styles, sections, and corporate formatting encoded as repeatable DOCX generation. Open Design lists the skill for discovery during planning; the full MiniMax AI skills repository carries scripts, references, and assets you copy into your agent skills directory before invoking by name or frontmatter triggers. Use it during validation when packaging a scope doc, during build when shipping internal specs or handoff docs, and at launch when attaching formal one-pagers to a site or email campaign. Intermediate complexity reflects OpenXML concepts and upstream bundle setup rather than a single chat prompt.794installs63Fal Image Editfal-image-edit advertises an agent-navigable path to AI-powered raster editing through fal.ai hosted models. Solo builders use it when they need quick inpainting, clean product cutouts, unwanted object removal, or stylistic passes on hero images without standing up their own GPU stack. The Prism listing is an Open Design catalogue entry: it helps your agent recognize intent during planning and points you to the fal-ai-community skills repository for the complete skill package. Expect API keys, network access, and upstream install steps when you move from discovery to execution. It sits in Build integrations because the value is connecting a production or marketing image pipeline to a managed inference endpoint, which is typical for indie SaaS and content sites polishing assets before validate-or-ship milestones.793installs64Gsap Scrolltriggergsap-scrolltrigger is the open-design agent skill for GreenSock ScrollTrigger—scroll-linked tweens and timelines, pinned sections, scrubbed progress, and parallax-style motion on landing pages and product UIs. Solo builders who want polished marketing sites without picking a library from scratch get a MIT-licensed, prototype-mode recipe that starts with gsap.registerPlugin(ScrollTrigger) and progresses through basic trigger wiring. The skill explicitly defers core tween APIs to gsap-core and gsap-timeline, React cleanup to gsap-react, and ScrollSmoother-style additions to gsap-plugins, which keeps responsibilities narrow. Use it when users mention ScrollTrigger, scroll animations, or pinning, or when they request scroll-based animation without naming a framework.793installs65Screenshots Marketingscreenshots-marketing helps solo builders produce marketing-grade product screenshots using Playwright, without juggling five manual capture tools every launch week. It is curated from Shpigford’s screenshots project and advertised in Open Design so your agent can plan capture jobs during distribution work—hero images for a landing page, framed shots for App Store listings, or polished frames for changelog posts. The catalogue entry is not the full runner by itself; you install the upstream repository into your skills directory and invoke screenshots-marketing by name or triggers like marketing screenshot and playwright screenshot. Intermediate complexity reflects browser automation setup, viewport discipline, and repeatable capture scripts rather than one-off OS screenshots. Pair it with a working staging URL or local dev server and consistent UI states so outputs look trustworthy on SEO landing pages and social previews.793installs66Figma Generate DesignFigma Generate Design is an Open Design catalog skill that helps solo builders bridge implementation and visual design by generating or updating Figma screens from existing code or plain descriptions, grounded in design-system components and tokens. It does not replace hands-on Figma craft; it advertises the upstream Figma skills workflow so your agent can find the capability during planning and you can install the full bundle with scripts and references from the official skills repo. Use it when you want parity between what you built in the app and what designers or stakeholders review in Figma, or when you need quick screen drafts that respect tokens instead of one-off rectangles. The entry is intentionally integration-forward: invoke by name or trigger phrase, then follow install steps for the upstream package. Best for teams-of-one who still want a single source of truth across code and design files during the build phase.792installs67Minimax Pdfminimax-pdf is an Open Design catalogue skill that advertises MiniMax’s PDF workflow for solo builders who need branded PDFs, e-guides, and reports without rebuilding typography and cover treatment each time. The local SKILL.md is a discovery stub: it explains what the upstream skill does—generation, fill-in, and reformat with a design-token system and fifteen cover styles—and tells you to install the full MiniMax-AI/skills bundle before running scripts and references. Use it during build when documentation, lead magnets, or customer-facing PDFs must match a consistent visual system. Complexity sits at intermediate because you coordinate upstream install paths, agent invocation by name, and document content. It is not a generic markdown-to-PDF one-liner; it is a curated document generator pattern aimed at design-system fidelity.792installs68Swiftui Designswiftui-design is a catalogued SwiftUI frontend design skill curated from the wholiver upstream bundle. Solo and indie builders use it when iOS or Apple-native UIs start looking like generic AI output and need a deliberate design direction, brand-consistent assets, and a structured critique pass. The workflow combines anti-slop guardrails, a design direction advisor, a brand asset protocol, and a five-dimension review so agents produce interfaces that feel intentional rather than template-driven. Install the upstream repository into your agent’s skills directory when you want the full references and scripts; invoke by name or trigger phrases such as swiftui design, ios design, or swiftui slop during planning and implementation. It pairs naturally with prototype and frontend build work and is especially relevant when you are polishing screens before review or launch.792installs69Figma Create Design System RulesFigma Create Design System Rules is an agent skill from the Open Design catalogue that helps solo builders and small teams generate project-specific design system rules for Figma-to-code workflows. Instead of scattering token names, component boundaries, and lint expectations across chat threads, the skill guides you to document them in a single rules artifact agents and engineers can follow when implementing screens. It is curated from Figma’s official skills repository and pairs naturally with Figma MCP workflows during planning: trigger phrases like “figma rules” or “design system rules” surface the skill before you codegen. Prism lists the catalogue entry for discovery; for scripts and references you install the upstream bundle into your agent skills directory. Expect intermediate familiarity with design tokens and your stack’s component model. The skill does not replace visual design in Figma—it standardizes how designs become consistent code.791installs70Image Enhancerimage-enhancer is a catalogued agent skill that guides your coding agent to improve image and screenshot quality—higher apparent resolution, sharper edges, and cleaner detail—for work you show to users or stakeholders. Solo and indie builders reach for it when README shots look soft, Figma exports need a polish pass, or tutorial figures fail the squint test on retina displays. The Prism entry surfaces discovery and planning triggers; the full procedural workflow, scripts, and references live in the ComposioHQ upstream bundle you install into your agent skills directory when you want end-to-end execution. It fits the Build phase documentation lane rather than a one-off chat tweak, and pairs naturally with other Open Design media skills when you are assembling launch-ready visuals without opening a separate desktop editor for every asset.791installs71Imagenimagen is an Open Design catalogue skill that routes solo builders to Google Gemini-powered image generation for UI mockups, icons, illustrations, and other visual assets. It is curated from sanjay3290’s upstream imagen project and surfaced during agent planning so you can discover the capability before installing the full skill bundle with scripts and references. Typical use spans validating a product look with quick mockups, filling frontend screens with placeholder art that feels on-brand, and producing simple icons without opening a separate design tool. The listing itself documents triggers and install steps; production runs require copying the upstream repository into your agent skills directory and invoking imagen by name or trigger phrase. It fits indie SaaS, content sites, and light game UI work where speed and iteration matter more than a formal design handoff.791installs72Figma Usefigma-use is the foundational Open Design catalogue skill for Figma Plugin API work. Solo and indie builders use it when they need agent-guided canvas writes, layer inspections, variable updates, or design-system maintenance without leaving their coding agent. The Prism entry advertises triggers and points to Figma’s upstream skills repository so you install the full bundle—scripts, references, and MCP-aligned workflows—into your agent skills directory. It is explicitly labeled as the prerequisite for sibling Figma skills in the same catalogue, which makes it the right first install before specialized Figma tasks. Ideal when you already use Claude Code, Cursor, or Codex for implementation and want repeatable Figma automation instead of manual file hopping. The README is a discovery stub; execution depth lives in the upstream package after install.790installs73Gsap Reactgsap-react is an agent skill curated from GreenSock’s official GSAP skills for React and React-based stacks such as Next.js. Solo builders install it when they need tweens, timelines, or scroll-driven motion inside components without memory leaks or SSR footguns. The skill steers agents toward @gsap/react’s useGSAP() hook instead of hand-rolled useEffect blocks, because the hook supplies scope and context-aware cleanup automatically. It documents installation, ref patterns, gsap.context() boundaries, and when to escalate to companion skills for core tweens, timelines, ScrollTrigger, or non-React frameworks. Use it during active UI implementation or when reviewing animation PRs; it does not replace scroll-specific or vanilla GSAP skills when those are the primary ask.790installs74Venice Audio Musicvenice-audio-music is an Open Design catalogue skill that connects solo builders to Venice.ai music generation—queueing jobs, polling completion, and fetching audio for short-form use cases. It targets jingles, ambient loops, and scratch scoring while you are still prototyping a game, content product, or SaaS demo, so you avoid hunting stock libraries for every iteration. The Prism listing is a planning-time advertisement: triggers and category metadata help your agent pick the skill, then you install the upstream veniceai/skills package for scripts, API details, and references. Expect API credentials and network access once the real bundle is in place. It complements visual and copy skills by closing the gap on cheap, rights-flexible audio for validate-and-build cycles rather than polished master tracks for commercial release.790installs75Youtube Clipperyoutube-clipper is an Open Design catalogue skill that points agents to automated YouTube clip workflows—ingesting source video, cutting highlights, adding captions, and exporting finished clips. It targets solo builders and creators who already publish long-form on YouTube and need shorts, teasers, or social cuts without living inside a nonlinear editor. The listing makes the capability discoverable during agent planning; the full editing automation lives in op7418’s Youtube-clipper-skill repository, which you install into your agent skills directory before invocation. Expect network access to fetch source material and filesystem writes for exports. It bridges Grow content operations and Launch distribution when you ship campaign clips alongside a feature or episode drop.790installs76Fal 3dfal-3d is an Open Design catalogue skill from the fal.ai community that advertises 3D model generation from text prompts or source images. Solo builders use it when they need fast spatial concepts—game props, AR preview meshes, ecommerce product sculpts, or mood-piece assets—without booting a full DCC pipeline on day one. The Prism listing helps your coding agent discover the skill during planning; executing the complete workflow requires installing the upstream fal-ai-community skills bundle and invoking fal-3d by name or listed triggers. It pairs naturally with validate-and-build loops where you prove a visual direction, then hand exported meshes or glTF to your engine, storefront, or AR prototype. Complexity stays at the integration layer in this stub: API usage, fal account setup, and export formats live in the upstream repository documentation.789installs77Paywall Upgrade Cropaywall-upgrade-cro is an agent skill curated from Corey Haines’s marketingskills collection for solo SaaS founders who need disciplined conversion thinking on monetization UI—not just prettier modals. It helps you frame upgrade screens, paywalls, and upsell moments with CRO-oriented structure so experiments are intentional rather than one-off copy tweaks. Invoke when you are designing a first paywall, refreshing pricing presentation inside the app, or planning A/B variants on upgrade paths. The Open Design catalogue entry surfaces triggers during planning; the full playbook, examples, and references live in the upstream repository you install into your agent’s skills folder. It complements pricing strategy work earlier in the journey and frontend implementation in Build, but its sweet spot is improving conversion on surfaces users already see during onboarding, limits, or plan changes.788installs78Ai Music AlbumAI Music Album is an Open Design–listed agent skill that advertises end-to-end AI music album production: thematic concept, lyric drafting, track sequencing, and export-oriented handoff. Indie builders and solo creators use it to structure an album experiment or a brand soundtrack without hiring a full production team on day one. The Prism entry curates discovery and planning triggers; the executable depth lives in the upstream bitwize-music-studio Claude AI music skills repository, which you install into your agent’s skills directory when you want original scripts and references. It sits earliest in Validate when you are proving a sonic direction, and extends into Build when you refine tracks and Grow when sound becomes part of content and lifecycle. Expect creative guidance and sequencing discipline rather than guaranteed commercial mastering or platform distribution in this catalogue slice alone.787installs79Domain Name BrainstormerDomain-name-brainstormer is a catalogue skill for solo founders at the very start of a project who need memorable names and a quick read on whether matching domains exist. It is positioned in the Idea phase as a discover-step helper: you describe your product concept or trigger phrases like “brainstorm domain” or “startup name,” and the agent follows the upstream ComposioHQ workflow after you install the bundle from awesome-claude-skills. The advertised behavior spans creative ideation and multi-TLD availability checking, which reduces the trap of falling in love with a name that is only purchasable at a premium. It is lightweight compared to full brand strategy—best paired with later validate work on positioning and landing copy. Install the upstream package into your agent skills directory, then invoke by skill name for structured output rather than one-off chat brainstorming.787installs80Fal Video EditFal-video-edit is a Prism catalogue skill that helps solo builders discover and invoke AI video editing through fal.ai’s hosted models when they already have footage to transform. It does not embed the full upstream workflow in the stub; instead it documents what the integration does—style remix, upscaling, background removal, and audio addition—and links to the fal-ai-community skills repository for scripts and assets. During build, you install the upstream bundle, then trigger the skill by name or with phrases like “video upscale” or “video style transfer.” It fits indie founders shipping marketing clips, in-app generative features, or automated content pipelines without running local GPU stacks. Expect agent-led planning plus network calls to fal rather than a single static template file.787installs81Fal TryonFal Tryon is an agent skill for solo builders and small ecommerce teams who want virtual try-on imagery without standing up custom CV infrastructure. It advertises fal.ai hosted try-on models for seeing how clothes look on a person, aimed at product pages, lookbooks, and styling experiments. The Open Design entry is a discovery and planning layer: frontmatter triggers help agents match user intent, while the body points to the upstream fal-ai-community skills repository for full scripts, assets, and references. You install the upstream bundle locally, then invoke by skill name or natural trigger phrases. Expect network calls to fal.ai and API credentials as part of the real workflow once upstream assets are present. It accelerates creative iteration during build and validate prototypes; it is not a replacement for merchant photo guidelines, model release compliance, or production CDN asset pipelines.786installs82Figma Generate Libraryfigma-generate-library is an agent skill for solo builders and small teams who maintain a Figma design system alongside a real product codebase. It advertises a curated workflow derived from Figma’s upstream skills repository, focused on generating or updating library structure, components, and variants so designers and agents do not drift from production UI. Use it when you are starting a design system from existing React or web components, after a refactor that renamed or split components, or on a recurring sync pass before a release. The catalogue entry points discovery during planning; running the full pipeline requires installing the upstream bundle and invoking the skill by name or trigger phrases. It pairs naturally with component-driven frontends and reduces hand-off friction between engineering and design without replacing human design judgment on tokens, accessibility, or brand voice.786installs83Gif Sticker Makergif-sticker-maker is a curated MiniMax-team skill surfaced in Open Design so agents can discover image-generation workflows during planning. Solo builders use it when they want branded, playful animated stickers from photos without hand-animating frames—useful for Discord, Slack, WeChat-style packs, or landing-page mascots. The catalogue entry points to the full upstream bundle on GitHub for scripts, assets, and references; locally you install that bundle and invoke gif-sticker-maker or its frontmatter triggers. It assumes network access to MiniMax’s API and fits builders who already treat generative media as part of their growth and community surface area rather than as in-app core logic.786installs84Gsap Timelinegsap-timeline is an official GreenSock-aligned agent skill for building ordered animation sequences with GSAP. Solo and indie builders use it when a landing page, product tour, or interactive demo needs several elements to move in a defined order—starts, overlaps, and pauses—without fighting ad-hoc setTimeout chains. The skill walks through creating a timeline, appending tweens, and placing each tween with the position parameter so motion reads as choreography rather than a pile of independent animations. It also covers nesting timelines for reusable scenes and controlling playback when users trigger or scrub motion. Pair it with gsap-core for single properties and eases, gsap-scrolltrigger when scroll drives the same sequence, and gsap-react when timelines live inside React components. Triggers match common agent prompts around animation timelines, sequencing, and motion choreography in GSAP or comparable libraries.785installs85Stitch Loopstitch-loop is an agent skill curated in Open Design from Google Labs Stitch for builders who feel the gap between Figma briefs and shipped UI. It describes an explicit iterative feedback loop: critique the current build against the brief, adjust implementation or design decisions, then ship a tighter increment—repeated until fidelity is acceptable. In Prism’s journey model it shelves under Build → frontend because the skill’s value is tightening what users see in the browser or app shell, though the same loop helps after validate prototypes or before launch polish. The packaged SKILL.md is primarily a discovery and routing layer: it points to the upstream repository for scripts, references, and the full workflow assets agents need to execute end to end. Solo founders using Claude Code, Cursor, or Codex can invoke it by name or listed triggers during planning so the agent chooses a structured iteration ritual instead of ad-hoc pixel nudging. Expect to clone or install the upstream Google Labs skills bundle when you want more than the catalog synopsis.785installs86Figma Code Connect Componentsfigma-code-connect-components is an Open Design catalogue skill that tells your coding agent how to discover and invoke Figma Code Connect workflows so design-system updates flow from Figma into the repository. Solo builders shipping branded SaaS or mobile web UIs often lose hours re-implementing component tweaks; Code Connect reduces that drift by declaring explicit links between Figma components and their code counterparts. This entry curates triggers and positioning from Figma’s upstream skills repository rather than re-embedding the entire MCP guide, so planning conversations surface the right skill name early. During Build, use it when you are standing up or extending a shared component library; it also helps Validate when a prototype in Figma must prove parity with a coded stub. Installation notes point you to clone or copy the upstream bundle into your agent skills directory before expecting scripts and references to run. Pair it with your Figma MCP setup and treat upstream README paths as the source of truth for exact connect files and repo layout.784installs87Ppt KeynoteKeynote-style Slides is an agent skill packaged as an Open Design template for builders who need polished presentation HTML without opening Keynote or PowerPoint. It instructs agents to produce Apple Keynote–quality layouts where every screen is one card: extreme simplicity, strong typography, and optional data visuals, all within a fixed 1280 by 720 frame suitable for screen sharing or hosted preview. The skill pairs with the html-anything workflow so you can iterate in the browser with debounced reload rather than exporting static PDFs on every tweak. Solo founders use it when validating a narrative on a landing call, launching a product tour, or repurposing marketing copy into a swipeable deck for investors and early users. It expects real content and data—explicitly avoiding lorem ipsum and placeholder images—so the output is cite-ready for AEO-friendly product pages. It is a visual template skill, not a slide CMS, analytics platform, or automated video renderer.784installs88Fal Restorefal-restore is a Prism catalogue skill that points solo and indie builders to fal.ai community restoration workflows for deblurring, denoising, face repair, and old-document recovery. It is aimed at agents in Claude Code, Cursor, Codex, and similar environments when marketing shots, user uploads, or archival scans need quality fixes without standing up local super-resolution stacks. The Open Design entry surfaces triggers and category metadata during planning; you install the upstream GitHub bundle into your agent skills directory to run the full workflow with original assets and scripts. Use it when hosted inference fits your budget and compliance posture, and pair it with your existing asset pipeline or CMS import steps. The skill does not replace a DAM or CDN—it routes restoration intent to the right community skill name and upstream source.783installs89Shader Devshader-dev is an Open Design catalog skill that advertises MiniMax-AI’s GLSL shader workflow so your coding agent can plan ray marching, fluid simulation, particle systems, and procedural generation for striking hero visuals and motion stills. Solo and indie builders shipping landing pages, SaaS dashboards, or interactive demos use it when stock imagery feels flat and you want distinctive procedural motion without becoming a full-time graphics engineer. The Prism entry does not embed the full upstream scripts; you install the bundle from the linked repository and invoke shader-dev by name or triggers during planning. It fits the Build phase when you are implementing frontend experiences, and it also supports earlier prototype mood-setting and later launch-ready visual polish when those assets are shader-driven.783installs90Sorasora is an Open Design catalogue skill that advertises OpenAI’s Sora video-generation capability to coding agents during planning, so solo builders can name a concrete workflow for short clips instead of vague “make a video” requests. It describes generating, remixing, and managing brief cinematic or b-roll footage via the Sora API—useful when you are validating positioning, storyboarding a landing hero, or mocking app promos before investing in a full editor pipeline. The skill does not embed the entire upstream repo; it tells you to install the OpenAI skills bundle and invoke by name or frontmatter triggers. Best for indie creators and product marketers iterating visuals quickly. Intermediate complexity because you need API access and clear creative briefs. Pair with distribution skills later when clips are ready for Launch.783installs91Venice Image EditVenice-image-edit is a Prism/Open Design discovery entry for the Venice.ai team’s image workflow: edits, upscaling, and background removal through their API. Solo builders who ship landing pages, app store shots, or in-product visuals use it when they want the coding agent to know this capability exists before implementation, then install the full upstream skill package from Venice’s GitHub repo so scripts and references match production. The local SKILL.md is intentionally thin—it advertises triggers and points at the bundle rather than replacing upstream assets. Pair it with generative or layout skills when you need pixels on the page; this skill is the Venice-specific integration hook, not a general design system. Intermediate complexity because you manage API keys, upstream install paths, and agent skill directories.783installs92Video Downloadervideo-downloader is a Prism catalogue skill that helps solo builders and indie hackers fetch videos from YouTube and additional platforms for offline editing, demos, or archival. Agents discover it during planning via Open Design metadata aligned with video-generation workflows; executing the full capability requires installing the ComposioHQ awesome-claude-skills upstream package into your active agent skills directory. It fits builders who create course content, social clips, or internal reference libraries without manually juggling browser extensions for every grab. Triggers are explicit in frontmatter so you can ask by skill name or natural phrases like youtube download. Respect platform terms of service and copyright where you operate—the skill automates retrieval mechanics, not licensing. After install, your agent can select formats and quality per upstream documentation.783installs93Fal Realtimefal-realtime is an Open Design catalog skill that surfaces the fal.ai community workflow for real-time, streaming AI image generation. Indie builders use it when they need fast visual iteration—moodboards, layout directions, and draft variations—without waiting on batch renders or manual Photoshop loops. The Prism listing helps your agent discover the skill during planning; you still install the upstream fal-ai-community/skills bundle to run scripts, credentials, and references. It shines in Validate when you are proving brand and UI direction, and it remains useful in Build when you are generating placeholder or exploratory art for landing pages and product mockups. Streaming behavior is the differentiator versus one-shot image APIs, making it a strong fit for conversational creative sessions with your coding agent.782installs94Fal Trainfal-train is an Open Design catalogue skill that tells your agent how to find and run custom LoRA training on fal.ai for personalized image generation aligned to a brand, character, or visual style. The Prism entry is intentionally thin: it advertises the capability during planning and directs you to install the upstream fal-ai-community skills repository for the full workflow, scripts, and references. Solo builders use it when generic diffusion models miss consistent identity or art direction and they already plan to host training on fal. After discovery, you inspect the upstream README, install the bundle into your agent skills directory, and invoke fal-train or natural trigger phrases. It does not replace reading fal training docs or curating a quality image dataset.782installs95Fal Upscalefal-upscale is an Open Design catalog skill for the fal.ai community bundle that upscales and enhances image and video resolution using hosted super-resolution models. Solo builders reach for it when moodboard or generation outputs are directionally right but too small or soft for production landing pages, app stores, or paid social. The Prism page helps agents discover the skill during planning; full execution depends on installing fal-ai-community/skills and configuring API access. It naturally follows exploratory generation: once you pick a winner frame, upscale before you Ship or Launch. It is less about ideation and more about production-quality deliverables, which keeps it anchored in Ship while still supporting Launch distribution assets and late Build polish when you freeze marketing imagery.782installs96Fal Kling O3fal-kling-o3 is an Open Design catalogue skill that surfaces fal.ai’s Kling O3 family—Kling’s high-end image and video models—so solo builders can ask their agent for generative visuals during planning. It fits indie founders shipping SaaS or content products who need short-form video, hero imagery, or iterative creative without maintaining a custom fal integration in every session. The Prism entry documents triggers and discovery; executable workflows, assets, and API details remain in the upstream fal-ai-community/skills repository you install locally. Use it when launch calendars need fresh creative fast, understanding that cost, quotas, and content policy live at fal.ai. After install, invoke by skill name or the listed trigger phrases so the agent routes to the upstream procedure rather than guessing model endpoints.781installs97Figma Create New Filefigma-create-new-file is an Open Design catalogue skill that tells your agent how to begin Figma work the right way: spin up a blank Design or FigJam file before components, workshops, or libraries pile on. Solo builders use it when they are still proving UX direction—empty file, then frames—or when they are formalizing a design-system run that expects MCP-guided Figma steps. The readme is intentionally thin locally: it points to Figma’s upstream skills repo for scripts and references, and positions this entry so planning-phase agents discover the capability by name or trigger phrases. Treat it as the front door to Figma MCP workflows, not the entire implementation skill pack. After the file exists, continue with downstream Figma skills from the same upstream bundle for variables, components, or reviews. Install upstream assets when you need executable scripts beyond catalogue discovery.781installs98ReplicateReplicate is an agent skill that helps solo builders discover, compare, and execute AI models through Replicate’s API, with a strong fit for image, audio, and video generation where model choice changes often. The Open Design entry advertises the capability during planning; the full upstream workflow, scripts, and references live in the replicate/skills repository on GitHub. Install that bundle into your agent’s skills directory when you need the complete implementation, then invoke the skill by name or using its frontmatter triggers. It suits indie teams shipping generative features in SaaS or content products without operating their own inference stack, and agents that need a stable handoff point when brainstorming which hosted model to call next.781installs99Venice Audio SpeechVenice-audio-speech is an Open Design catalogue skill that tells your coding agent how to use Venice.ai for text-to-speech when you build products that need spoken output. Solo and indie builders shipping assistants, course content, or lightweight media apps can invoke it during planning so the agent picks correct models, voices, formats, and streaming patterns instead of guessing API shapes. The Prism entry is intentionally discoverable: frontmatter lists triggers like voiceover and narration, and the body explains that full upstream workflows live in the Venice skills repo you install beside your agent. Use it when audio is a first-class UX requirement; skip it when you only need static UI copy. After discovery, install the upstream bundle and ask the agent to run the skill by name or trigger phrase for end-to-end implementation help.776installs100Venice Image GenerateVenice Image Generate is an Open Design catalogue skill that points solo builders at Venice.ai’s image API capabilities—endpoints, styles, and invocation patterns—so coding agents can plan generative visuals without hunting docs mid-session. The Nexu entry is intentionally thin: it advertises the skill for discovery during planning and tells you to install the upstream Venice.ai skills bundle for complete assets and scripts. Use it when you need on-brand or exploratory images for landing pages, app shells, or content products while building. It is an integration-oriented generative media skill, not a hosted image editor. After catalogue match, clone or sync from the upstream GitHub repo and invoke by skill name or listed triggers.776installs101Pixelbin Mediapixelbin-media is an Open Design discovery skill that connects solo builders to Pixelbin’s upstream agent bundle for generative and transformative media at web scale. Rather than embedding every API script inline, it advertises the capability during planning so Claude Code, Cursor, or Codex can invoke pixelbin-media or natural trigger phrases, then directs you to install the full upstream repo where assets and references live. The positioning—85+ APIs spanning generation, editing, and CDN-oriented delivery—suits SaaS marketing sites, content products, and storefronts that need on-brand visuals without a dedicated design hire. Use it in Build when composing landing sections, hero assets, or automated thumbnail pipelines, and again in Launch when refreshing SEO pages or campaign creatives. Validate prototypes can lean on the same triggers for quick visual proof. Treat this Prism entry as the map; execution depth depends on installing the Pixelbin skills package named in the readme.775installs102Faq PageFAQ Page is an open-design agent skill that produces a support-oriented FAQ prototype when a brief asks for FAQ, help center, Q&A, or support page. Solo builders validating SaaS, ecommerce, or service sites invoke it during Build to turn DESIGN.md component tokens into a coherent desktop HTML page with accordion interactions, search, and category pills. The workflow forces realistic domain copy (roughly a dozen to eighteen entries), handles thin content by simplifying layout, and keeps craft bars for typography and accessibility. It is a page generator, not a CMS integration—ideal before handoff to your stack or for quick stakeholder previews in Validate when a landing needs a credible support section.774installs103Slack Gif CreatorSlack GIF Creator is a catalogue skill that helps agents plan and invoke workflows for Slack-ready animated GIFs. Solo builders use it when they want on-brand or playful motion assets for Slack channels without manually tuning frames in external tools. The Open Design entry documents upstream provenance from Anthropic’s official skills repository and explains that full scripts and references live in the upstream bundle—install that bundle into your agent skills directory for the complete workflow. At a high level, the skill emphasizes validators for Slack constraints and composable animation building blocks rather than one-off huge exports. It fits builders who already use Claude Code or similar agents for creative chores and need discoverable, named invocation during planning.772installs104Venice Videovenice-video is an Open Design catalogue skill that points solo builders and agents at Venice.ai’s official skills bundle for video generation and speech-to-text style transcription over the Venice API. Prism lists it so planning agents can discover the capability by name or trigger phrase before committing to an upstream install. The local SKILL.md does not embed the full workflow; it documents what the integration does, where the authoritative package lives, and how to clone or copy that bundle into an agent’s skills folder. Use it when you want generative video or transcribe flows in a Claude Code or Cursor stack without manually searching skills.sh or GitHub. After install, invoke venice-video or the documented triggers and follow upstream README paths for credentials, endpoints, and asset handling.771installs105Wpdswpds is a WordPress Design System agent skill that steers solo builders and small teams toward WordPress’s official design tokens, typography, and component patterns when building or refreshing themes and sites. The catalogue entry surfaces the capability during agent planning via Open Design; full procedural assets live in the upstream WordPress/skills repository, which you install into your agent’s skills directory when you want the complete workflow. Use it when conversations mention wpds, WordPress design, WP tokens, or the WP design system—especially before UI commits on block themes, admin-adjacent surfaces, or marketing pages that should feel native to WordPress. It reduces one-off color and spacing choices that fight core patterns and helps agents reason about consistency with Gutenberg-era expectations. Pair it with your normal frontend build loop; it is a design-system integration skill, not a hosting or plugin installer.767installs106Data Reportdata-report is a Nexu Open Design template skill that helps solo builders ship credible data stories fast. You feed structured inputs—CSV exports from Stripe, product analytics, or finance spreadsheets—and the agent assembles a finance-scenario desktop report with KPI cards, multi-chart sections, and tabular detail using Chart.js or ECharts loaded from CDN. The template encodes visual signature rules (fixed-height chart wrappers, 3–5 KPI slots, minimum two charts) so outputs look investor-ready instead of generic dashboard clones. Use it when validating whether metrics resonate before you wire a live backend, when drafting weekly growth memos for Grow, or when you need a one-off HTML artifact for Launch distribution. It pairs with the html-anything upstream lineage: preserve template structure, use authentic numbers, and avoid placeholder imagery so the page can ship as a standalone file or embed.744installs107Poster Heroposter-hero is an agent skill from the open-design template family that turns your real marketing copy into a vertical poster or WeChat Moments–style share image with strong visual impact. Solo and indie builders use it when they need a launch announcement, promo slide, or social asset without opening Figma or hiring a designer. The skill encodes a concrete Tailwind-style layout—full-height gradient mesh background, oversized headline, short subtitle, a small set of icon-backed bullets, and a brand or QR placeholder in the corner—so the agent preserves a consistent visual signature while substituting your product name, offer, and proof points. It pairs with the html-anything upstream workflow and optional design-system flags, emphasizing prototype HTML surfaces and live preview rather than static mockups. Use it during Launch when you are ready to distribute, and again in Grow when you need fresh campaign creatives, as long as you feed authentic content and data instead of filler text.737installs108Deck Guizang EditorialGuizang Editorial E-Ink Deck is a design-forward agent skill that formats your messaging into an editorial magazine aesthetic with electronic-ink restraint. Solo builders and indie marketers use it when they need a credible horizontal deck for product narrative, chaptered storytelling, or metric-led slides without default corporate templates. The skill ships ten layout patterns and five curated color systems, with bilingual metadata and ties to the guizang-ppt lineage and nexu-io html-anything preview flow. Invoke it when you have substantive bullets, numbers, or chapter structure and want the agent to preserve the template’s serif display and paper-like rhythm. It is not a code generator or analytics pipeline; it is a structured visual template layer so launch and growth content looks intentional before you record, post, or pitch.719installs109Web Design Guidelinesweb-design-guidelines is an agent skill that surfaces Vercel engineering web design standards for solo builders polishing SaaS and marketing product UI. It is aimed at developers and indie founders who want consistent layout, typography, color, motion, and accessibility without reinventing a design system from scratch. Use it during frontend implementation, prototype refinement, or pre-launch UI passes when you ask for Vercel design patterns, a design checklist, or product UI standards. The Open Design entry advertises the capability for planning and discovery; the full upstream skill bundle on GitHub carries the complete workflow, scripts, and references your agent should install into its skills directory when you need executable depth. Treat it as normative UI guidance that pairs well with component libraries and accessibility review, not as a code generator on its own.712installs110Card Xiaohongshucard-xiaohongshu is an open-design agent skill that formats your real content into Xiaohongshu-style knowledge cards arranged as a swipeable multi-card carousel. Solo and indie builders use it when they want polished 图文 without rebuilding layout in Figma or a separate design tool—especially for habit lists, tool roundups, and teachable threads aimed at 小红书 or WeChat Moments. The skill enforces a visual signature: Morandi-style gradients, strict 1080×1440 frames, a high-impact cover, one idea per middle card, and a summary CTA on the last slide. Card count scales with information density rather than filler. It pairs with the html-anything prototype flow so agents generate live HTML you preview in the browser, using authentic copy and data instead of lorem ipsum or placeholder images. Best when you already have substance and need channel-native packaging fast.708installs111Article MagazineArticle-magazine is an open-design agent skill that packages a magazine-style article template inspired by Huashu and huashu-md-html. Solo builders use it when they already have Markdown or notes and need a credible long-form HTML essay for a landing story, newsletter issue, or public post without rebuilding typography and section rhythm from scratch. The skill emphasizes preserving the template’s visual signature while swapping in real copy, links, and data. It sits in the Launch phase as a distribution artifact generator rather than a generic site scaffold: one focused long page meant to read like editorial design on desktop web. Pair it with your writing workflow in Validate or Grow, then invoke when you are ready to ship the piece as styled HTML. Complexity stays beginner-friendly because the contract is template fidelity plus content replacement, not a full design system build.702installs112Resume ModernResume Modern is a design skill from Open Design that formats a solo builder’s career story as a modern minimal CV on one A4 page. It encodes layout rules—millimeter-sized container, two-column structure, typography hierarchy, and contact strip—so your agent outputs HTML or markdown that matches a fixed visual signature instead of generic résumé fluff. The template expects real companies, dates, and metrics; it explicitly avoids lorem ipsum and placeholder imagery. Use it when you need a polished PDF or print resume for investors, clients, or hiring while you are still running everything yourself. It pairs with upstream html-anything style prototyping and desktop web preview. It is not a job-matching or ATS-scoring tool; it is a consistent visual template you fill with authentic content.697installs113Deck Open Slide Canvasdeck-open-slide-canvas is an Open-Design agent skill for building presentation decks on a locked 1920×1080 canvas using React-level composition instead of one fixed template. Solo and indie builders use it when they need investor pitches, product walkthroughs, or conference slides that still feel designed—not generic slide masters—with real headlines, metrics, and imagery called out explicitly in the example prompt. The skill targets the design/slides scenario with optional i18n prompt variants and integrates preview via index.html so agents can iterate layouts in the browser. It pairs well with content you already have from validation or launch work and outputs a cohesive deck signature rather than lorem ipsum placeholders. Complexity is intermediate because you should be comfortable with React component structure and agent-driven HTML preview loops.695installs114Video Hyperframesvideo-hyperframes is an open-design agent template skill from the Hyperframes / Remotion ecosystem that helps solo builders script cinematic, continuous-frame product or workflow videos without starting from a blank timeline. It emits a sequence of structured sections (hyperframes) sized for 1920×1080 16:9, with autoplay-friendly behavior and metadata patterns illustrated by the bundled AI-workflow example (eight frames, progress bar, and frame metadata). The skill is invoked when you want agent assistance to turn substantive narrative, UI walkthroughs, or launch storyboards into HTML-first frame scripts you can feed Remotion or Hyperframes to export mp4. It preserves the template visual signature while insisting on authentic data and copy rather than filler, aligning with Nexu open-design preview and debounced HTML reload workflows for iterative polish before render.694installs115Mockup Device 3dMockup-device-3d is an open-design agent skill that helps solo builders produce marketing-grade device showcases without rebuilding 3D scenes from scratch. It targets the Device 3D Showcase template: a dual-screen iPhone and MacBook composition where each display carries live HTML rather than flat screenshots, aligned with hyperframes-style product storytelling. Invoked when you need catalog heroes, launch graphics, or pitch visuals, the skill steers the agent to keep the template’s refraction and 360-degree turntable look while replacing placeholder copy with your actual interface markup. It fits Prism’s journey as a bridge between what you built in the browser and what you ship outward—ideal for founders who already have index.html or component HTML and want a credible device frame for distribution channels. Complexity stays approachable because the surface is a fixed poster scenario with optional design-system bypass and desktop web prototype preview.690installs116Deck Swiss Internationaldeck-swiss-international is an Open Design deck skill that applies Swiss International typography and grid discipline to your content: a 16-column structure, a single bold accent from a small palette, and twenty-two fixed layouts so agents do not freestyle slide composition. Solo builders preparing launch narratives, product roadmaps, or investor snapshots use it when they need editorial credibility without hiring a slide designer. The template stresses factual KPI towers, full-bleed titles, and consistent accent usage across a 16:9 marketing deck rendered as web HTML with live preview. Featured/recommended metadata in the package signals it as a default marketing scenario within nexu-io’s design catalog. Beginner-friendly for operators who already have structured copy; the agent’s job is to map content into locked layouts while avoiding placeholder text and stock filler called out in the example prompt. Best paired with validated messaging from Validate or Grow content work, then shipped as a distribution asset.681installs117Doc Kami ParchmentDoc Kami Parchment is an Open Design skill that applies the Kami parchment document system: warm off-white canvas, ink-blue monochrome accents, one serif stack, and report-grade typography suited to one-pagers, letters, and long editorial pages. Solo builders use it when a landing PDF, founder letter, or studio issue needs a consistent literary look without hiring a layout pass for every draft. The template ships with featured ranking metadata, bilingual naming, and an example prompt that tells the agent to preserve the visual signature while substituting real data. Preview runs as HTML with debounced reload; design_system does not require a separate token package. It fits Validate when you need a credible one-page narrative and Build when documentation or release notes must match a premium editorial brand.676installs118Frame Light Leak CinemaLight-Leak Cinematic Frame is an agent skill packaged as a video-oriented HTML template for solo builders who need polished opening titles or chapter cards without opening a full video editor. It encodes a consistent cinematic signature: warm orange light leaks, grain, letterboxing, and bold serif typography so agent runs stay on-brand across reels or product walkthroughs. The skill is invoked when you already have copy or section labels and want them rendered as a film-style frame rather than a generic slide. It upstreams from html-anything style tooling and targets markdown or catalog-driven workflows where aspect ratio and surface type are declared up front. Builders shipping demos, course modules, or launch teasers use it in the Build phase to generate presentation assets that read as premium editorial video, while keeping implementation lightweight compared to timeline-based editing suites.671installs119Frame Data Chart NytNYT-Style Data Chart Frame is an Open Design agent skill that wraps your real numbers and narrative into a hyperframe with newsroom typography, accent color, and staggered chart reveals. Solo and indie builders use it when they need credible, publication-quality motion charts for product videos, investor updates, or growth storytelling without hand-building D3 or After Effects pipelines. The template expects authentic content and data rather than placeholders, and ships as an HTML video surface with catalog-aligned examples. It pairs naturally with html-anything upstream workflows when you already have structured copy or metrics and want a consistent editorial look across frames.670installs120Frame Logo OutroFrame Logo Outro is an agent skill backed by the Open Design / HTML Anything template stack for turning real brand content into video outros and closing frames. Solo and indie builders use it when a launch video, tutorial, or demo needs a consistent sign-off—logo pieces easing in, a controlled glow bloom, then tagline and optional call-to-action—without opening a full motion-design toolchain. The skill expects you to supply authentic copy and assets so the output reads as production-ready distribution material, not a generic mock. It fits the Launch phase when you are packaging clips for YouTube, product sites, or social, and pairs naturally with broader brand or generative-media workflows when you already have messaging and marks defined.669installs121Frame Flowchart Stickyframe-flowchart-sticky is a Nexu OpenDesign agent skill for the Sticky Flowchart Frame video template. Solo builders and small teams use it when they need a polished, whiteboard-style process diagram as a video surface rather than a static slide deck. The template combines SVG curve connectors, colored sticky-note nodes, and light cursor interaction so flows such as user onboarding or internal ops read like a facilitated brainstorm. Metadata points at html-anything upstream, HTML entry preview, and bilingual example prompts so agents preserve the visual signature while swapping in authentic steps and labels. It is categorized under video/operations scenarios and works with assistants that follow the Agent Skills spec. Reach for it when you already have the flow logic and want a presentable 16:9 frame without rebuilding diagram aesthetics by hand in a generic design tool.668installs122Frame Liquid Bg HeroLiquid Background Hero is a design skill that packages a recognizable fluid hero signature for solo builders who need a credible landing or video opener fast. It documents intent for SaaS tops, video head frames, and poster backdrops while degrading gracefully from WebGL aesthetics to CSS and canvas so one HTML file can open offline. The skill specifies canvas sizes, layered radial gradients with staggered keyframes, blend modes, and optional blur to mimic aurora-like motion without a full VFX pipeline. Agents follow the template block to swap in real messaging and brand colors rather than placeholders. It fits Prism’s Validate and Launch paths when you are testing positioning creative or shipping shareable visuals before investing in a design system. Pair it with your copy and analytics later; it is a visual scaffold, not a component library.667installs123Card TwitterTwitter Share Card is an open-design template skill that helps solo builders ship polished social visuals without opening Figma for every post. The agent follows a fixed layout spec: wide 16:9 frame, high-contrast hero line, attribution row, categorical chip (Insight, Data, Quote), and watermark, with optional dark or light treatment driven by content tone. It is categorized as marketing card prototyping on desktop web, backed by HTML preview workflows from the html-anything lineage. Use it when you have copy or a number worth amplifying on X and want a consistent brand look across a thread or launch week. The skill optimizes for screenshot-and-post velocity rather than automated API posting, so deliverables are ready-to-capture pages aligned to the template’s visual signature.666installs124Frame Glitch TitleFrame Glitch Title is an agent skill backed by the open-design / hyperframes catalog that applies a fixed digital-glitch title frame aesthetic to your own content. Solo and indie builders use it when they need a credible cyberpunk or signal-loss hero for a launch video, stream overlay, or product trailer without learning compositing tools. The skill encodes aspect ratio, emoji-led naming, and an example prompt that instructs the agent to keep chromatic offset and corruption motifs while replacing placeholder text with real headlines and metrics. It fits the Validate-to-Launch stretch when you already have messaging and want a repeatable visual template rather than one-off Figma frames. Prism lists it as generative media for distribution-ready video surfaces, with optional design-system hooks disabled so the frame stands alone.666installs125Vfx Text CursorVfx-text-cursor is an open-design agent skill template for solo builders who need cinematic opener clips without a motion designer on retainer. It encodes a specific visual language—hot pink and cyan chromatic cursor trails, directional flares, and staged typography reveals—for turning a real quote or thesis into a 16:9 video intro. The skill ships with bilingual metadata, a markdown example spec, and an HTML preview surface so agents preserve the template’s signature while swapping in your actual copy and data. It fits the Launch distribution shelf when you are preparing product announcements, course trailers, or social snippets where readability and spectacle must coexist. Builders using Claude Code or Cursor can follow the packaged example_prompt to avoid generic placeholders and keep pacing word-by-word. Treat it as a generative media pattern tied to the Hyperframes catalog rather than a general UI component library.664installs126Frame Macos Notificationframe-macos-notification is an open-design HTML template skill that helps solo builders and indie teams produce believable macOS notification banners for video overlays and launch teasers. Instead of mocking notifications in Figma or After Effects by hand, you feed the template your app icon, headline, and one or two lines of body copy and get a frosted-glass banner that reads like a native macOS alert on camera. It targets the hyperframes catalog flow: markdown-described examples, an index.html preview surface, and prompts that insist on real product data rather than placeholders. Use it when you are cutting a feature announcement reel, padding a Loom walkthrough with a “you’ve got a new update” moment, or stitching a short social clip that needs a polished OS-level visual without building a motion graphics stack. The skill does not replace a full design system; it is a focused generative template tuned for 1920×1080 video or a compact 480×120 strip. Intermediate familiarity with editing HTML/CSS in an agent-assisted repo is enough to ship usable overlays quickly.663installs127Social X Post CardSocial X Post Card is an open-design agent skill that maps your real tweet copy into a realistic X (Twitter) post card, complete with engagement stats and visual polish suited to video overlays or standalone image cards. Solo builders and indie marketers use it when they need launch-ready social creative without opening Figma for every variant. The template emphasizes dark-mode fidelity, metric rows, and aspect ratios common in short-form video and square feed posts. It plugs into the nexu-io html-anything style workflow with an HTML prototype surface, optional design-system bypass, and an example prompt that forbids placeholder text and fake imagery. Install it when you are shipping announcement threads, quote cards, or teaser clips and want consistent 𝕏 branding without hand-coding CSS each time.661installs128Docxdocx is an Open Design catalogue entry that advertises Anthropic’s upstream Word skill so your agent discovers document workflows during planning. It targets solo builders who need professional .docx deliverables—design briefs, copy documents, and comment-ready reviews—without manually fighting OOXML and revision markup. The local SKILL.md explains what the skill does and how to install the full upstream tree from GitHub when you need original assets and scripts; invocation is by name or frontmatter triggers such as “tracked changes” or “design brief doc.” Use it in Validate when scoping narrates in Word, in Build when specs and copy live alongside code, and in Ship when review packets need redlines. It is a skill package pointer, not a hosted document service; you still run the agent with filesystem access where upstream tooling expects it. Complexity is moderate because mastery depends on the installed upstream bundle rather than this stub alone.659installs129Social Reddit CardSocial Reddit Card is an open-design agent skill that renders a believable r/programming-style Reddit post as standalone HTML for marketing and storytelling. Solo builders use it when they want thread-like social proof or narrative hooks in launch content, short-form video overlays, or link previews without designing Reddit chrome from scratch. The skill keeps the template’s visual signature—dark mode, vote rail, and comment metrics—while you supply authentic title, body, subreddit, and counts. It ships as a categorized card template with optional i18n prompts, desktop web prototype settings, and a documented example (AITA-style story with 12.3k upvotes). Invoke it once your message is final enough to paste; the agent fills the layout and you export or screen-capture for distribution.658installs130Social Spotify Cardsocial-spotify-card is an open-design agent skill that produces a Spotify Now Playing-style card with album artwork, elapsed/total progress, and playback-control chrome for video overlays or personal homepages. Solo creators shipping landing pages, stream layouts, or portfolio sites install it when they want a recognizable music-player aesthetic without rebuilding Spotify UI from scratch. The template is categorized as a card scenario for personal use, ships with HTML preview at index.html, debounced reload for iteration, and explicit guidance to preserve the visual signature while substituting real titles, artists, and timing. It pairs with nexu-io’s html-anything lineage and hyperframes-style catalog examples. Use during Build when you are polishing frontend presentation or demo assets; hand off to Launch distribution when the card ships in a public URL or stream package.654installs131Weread Year In Review Video TemplateThe weread-year-in-review-video-template skill turns reading telemetry and bookshelf artifacts into a polished vertical video story suitable for WeRead, Goodreads, Readwise, Notion reading logs, book clubs, or personal learning recaps. Solo builders who want shareable annual content without a motion-design team can start from a structured HyperFrames composition: warm paper texture, editorial Chinese typography, book-page transitions, and scenes for active days, highlights, notes, keywords, and a closing persona card. Triggers in the skill mirror real user language—from English “reading year in review video” to 微信读书年度报告—so agents pick it when the brief is explicitly a recap template rather than a generic landing page. The open-design metadata defines a mobile 9:16 surface with HTML preview entry points and deterministic motion, which keeps renders repeatable for social distribution. Use it when you already have reading stats and copy and need a template-shaped deliverable, not when you need a live data pipeline from a reading API.324installs