vibe-building-skills

Orchestrates non-trivial software implementation with GPT-5.6 Sol as the decision-making root and bounded Sol Light/Medium, Terra, Luna, GPT-5.5, and Claude Sonnet 5 executors. Use when the user asks to build, implement, integrate, migrate, refactor, or fix work spanning multiple files or systems; requests parallel agents or an implementation plan; or needs evidence-backed execution with specialist routing, review gates, and end-to-end validation. Produces a grounded specification, dependency-aware task graph, implemented changes when authorized, and a verified completion report. Skip trivial one-file changes and explanation-only requests.

Use when the session model is Opus 4.8 (claude-opus-4-8) and the user asks to build, add, implement, refactor, migrate, or integrate something non-trivial — 3+ files, a new dependency or subsystem, an API contract or database schema change, or meaningful ambiguity or risk — or says "plan this out", "spec this", "how should we build", "create an implementation plan". Also use when the user says "just implement it" but the work is non-trivial. Do NOT use on Claude Fable 5 or any other model, nor for one-line fixes, typo corrections, single-function changes, or questions that require no code change.

Use when the user wants to RUN a sprint as a facilitated workshop with a virtual team — not just get sprint-style analysis. Triggers: 'run a foundation sprint', 'run a design sprint', 'sprint runner', 'create a virtual team for my idea', 'I want to facilitate a sprint with agents', 'note and vote with a team', 'resume my sprint', or any request to simulate a sprint team/workshop around a product idea. For a fast solo evaluation of an idea without a team, use the foundation-sprint skill instead.

Run an evidence-based product design and UI/UX review of a digital product (web, mobile, desktop) — heuristic evaluation, task walkthroughs, visual and interaction critique, and WCAG 2.2 AA accessibility checks. Use when the user says 'review my app/page/site', 'critique this UI', 'something feels off', 'why are users dropping off', 'audit our onboarding/checkout/dashboard', 'is this usable/accessible', or shares a URL, screenshot, Figma link, or codebase and asks what to improve — even if they never say 'design review'. Produces a Design Review Report: executive summary, prioritized findings table with evidence and severity, flow walkthroughs, accessibility blockers, and next steps routed to the skill that fixes each category.

Produces branded visual identity elements — logo concepts, icon sets, seamless patterns, brand illustrations, and textures — as generation-ready prompt specs, and generates them via FAL.ai fal-ai/nano-banana-pro when the automation is configured. Use when the user says 'design a logo', 'I need icons', 'brand pattern / texture', 'brand illustrations', 'visual identity elements', or 'make my brand assets'. Produces one prompt-spec block per asset (asset type, style, palette with hex, composition, negative constraints), a variant plan for user selection, file-naming conventions, and a brand-guidelines entry for each approved asset; runs docs/creative_cli.py brand when FAL_API_KEY is set.

Produces professional AI product photography — clean e-commerce shots, lifestyle scenes, hero images, macro detail shots, and flat lays — as generation-ready prompt specs, and generates the files via FAL.ai fal-ai/nano-banana-pro when the automation is configured. Use when the user says 'product photos', 'shots of my product', 'images for my store / listing / landing page', 'hero image for my product', 'lifestyle photos', or shares a product needing marketing imagery. Produces one prompt-spec block per shot (shot type, lighting, background, composition, aspect ratio, negative constraints) with a shot list, variant plan, and file-naming conventions; runs docs/creative_cli.py product when FAL_API_KEY is set.

Audit a frontend against WCAG and fix what fails: semantic HTML, keyboard navigation, focus management, screen-reader support (ARIA), color contrast, forms, and media alternatives. Use when the user asks 'is this accessible', 'make this WCAG/AA compliant', 'add keyboard navigation', 'fix screen reader support', 'a11y audit', 'we got an accessibility complaint/legal notice', or whenever building/reviewing UI components with interactive behavior — even if accessibility isn't mentioned. Produces a prioritized accessibility audit (issue × WCAG criterion × severity × fix) plus the implemented fixes in code.

Audit and refactor a frontend codebase's component structure, or design a component library from scratch, using atomic design, composition, and disciplined prop interfaces. Use when the user says 'my components are a mess', 'this component is 800 lines', 'refactor my components', 'build a component library', 'how should I structure my components', 'too much prop drilling', 'we keep rebuilding the same button', or asks where a new component should live — even if they never say 'atomic design'. Produces a component inventory audit, refactored component code (extracted atoms/molecules, typed prop interfaces, variant systems), and per-component documentation.

Translate a design (Figma mockup, spec, screenshot, or verbal description) into code with pixel- and motion-level fidelity, using the browser's rendering pipeline as a constraint: exact spacing/timing via tokens, GPU-safe animations, no layout thrashing, 60fps interactions. Use when the user says 'implement this design', 'make it match the mockup', 'it doesn't look like the Figma', 'the animation stutters/drops frames', 'set up design tokens', or when a diff between design and implementation needs to be closed. Produces the faithful implementation plus a fidelity report (design value → implemented value → any deviation and why) and performance notes.

Audit and implement an app's error UX end-to-end: rewrite error messages, place them correctly, add recovery actions, and build error states for validation, network, permission, system, and 404 failures. Use when the user says 'improve my error handling', 'my error messages suck', 'users get stuck when something fails', 'add form validation errors', 'handle offline/network errors', 'design a 404 page', or when a review finds bare 'Something went wrong' strings, silent failures, or errors with no way forward. Produces an error-state matrix (every failure mode × message × recovery action) plus the implemented error components and copy.

Design and implement the motion layer of a UI — microinteractions, transitions, hover/press feedback, and animation timing/easing — as working CSS/JS plus a written animation spec. Use when the user says 'the app feels janky', 'animations feel off', 'make it feel smoother/snappier', 'add a transition', 'the modal just pops in', 'buttons feel dead', 'add some polish/delight', or when interactions give no feedback. Also use to audit existing animations for timing, easing, performance (GPU properties), and reduced-motion support. Produces an animation spec (duration/easing per interaction class) and the implemented transitions.

Audit an app's async surfaces and implement the right loading, progress, and empty states — skeleton screens, spinners, progress bars, first-use/no-results/error-empty views — chosen by expected latency and layout knowledge. Use when the user says 'add loading states', 'the page just goes blank while fetching', 'add a skeleton screen', 'users think it's frozen', 'spinner or skeleton?', 'design the empty state', 'zero state', or when reviewing an app whose fetches show nothing. Produces a loading/empty-state audit of every async surface plus the implemented components.

Diagnose and fix how slow a frontend feels and measures: optimistic UI for mutations, latency masking (skeletons, blur-up, preloading, staggering), request batching/deduplication, and Core Web Vitals (LCP/INP/CLS) remediation. Use when the user says 'the app feels slow', 'clicks feel laggy', 'improve performance', 'my Lighthouse score is bad', 'fix my Core Web Vitals / LCP / CLS', 'make saves feel instant', or when interactions visibly wait on the network. Produces a performance audit (findings ranked by user impact, with measured or user-reported evidence) plus the implemented fixes — optimistic mutation code, masking treatments, and network-layer improvements.

Build or repair a color system: full brand and neutral ramps, semantic color tokens, contrast-verified pairings, dark mode, and status colors — delivered as a tokens file (CSS variables / Tailwind theme) with every text pairing's contrast ratio computed. Use when the user says 'the colors look off', 'colors clash', 'add dark mode', 'is this accessible', 'pick a palette', 'too many blues in the codebase', 'make it match our brand', or when contrast failures are found. Extracts the de facto palette from the code first, consolidates drift into ramps, chooses harmony deliberately, and verifies WCAG AA on every shipped pair in both modes.

Create or extract a design-token foundation — colors, typography, spacing, radii, shadows — as a real tokens file (CSS variables and/or Tailwind theme) plus a migration map from current hardcoded values. Use when the user says 'set up a design system', 'our styles are inconsistent', 'we have 50 shades of the same blue', 'add design tokens', 'formalize our design', 'make new features match', or when starting a new frontend that needs a foundation before building screens. Also the first step whenever other design work keeps fighting ad-hoc values. Produces the tokens file, a global→semantic token structure wired for dark mode, and a short list of the worst drift to migrate first.

Diagnose a codebase's frontend design maturity and route the work through the right sequence of frontend-design skills. Use when the user makes a broad or vague design request — 'make this look better', 'polish the UI', 'this feels like an MVP', 'our frontend is a mess', 'where do I start with design', 'make it feel professional', 'plan a design system' — or when a request spans multiple design concerns and it's unclear which single skill applies. Assesses maturity by inspecting the actual code (Tailwind config, CSS variables, component structure, hex-code drift), picks one of five implementation paths, and produces a design roadmap: an evidence-backed maturity diagnosis plus an ordered skill sequence with the artifact each step passes to the next. If the request clearly maps to one concern (only spacing, only color, only type), skip this and invoke that skill directly.

Fix or build responsive layout: page shells, grids, card layouts, sidebars, breakpoint strategy, container queries. Use when the user says 'this breaks on mobile', 'the layout is messy', 'things don't line up', 'our spacing is inconsistent between sections', 'make this responsive', 'center this properly', or asks for a dashboard/page/grid layout. Inspects existing CSS for layout smells (fixed widths, absolute positioning, media-query sprawl), then produces working layout code — refactored CSS/JSX with Grid/Flexbox decision rules applied, breakpoints derived from content, and a responsive verification checklist. Hands off pure spacing-token questions to design-foundation and text-readability questions to typography-system.

Build or repair a typography system: a modular type scale, font choices and loading, hierarchy, line-height/line-length, and fluid sizing — delivered as tokens (Tailwind fontSize config or CSS variables) applied to real components. Use when the user says 'the text looks off', 'fonts are inconsistent', 'headings don't stand out', 'pick fonts for me', 'it's hard to read', 'too many font sizes', or when an audit finds font-size sprawl. Chooses the scale ratio from product density (dense app vs marketing vs editorial), verifies contrast and measure, and produces the type tokens plus a migration map from existing rogue sizes.

Refactor cluttered or flat interfaces so users instantly see what matters, using size, weight, contrast, and whitespace — not more color. Use when the user says 'it looks cluttered', 'everything competes for attention', 'it looks amateur/flat', 'users can't find the button', 'make this scannable', 'this screen is overwhelming', or shares a screen where nothing stands out. Audits the actual component: lists every element's current visual weight vs intended priority, then delivers refactored component code (CSS/TSX) with a before/after hierarchy table explaining each change. De-emphasizes secondary elements before amplifying primary ones, and applies Gestalt grouping via spacing.

Facilitates an interactive session that turns a product or subject into an original, internally consistent visual design language for any artifact — infographics, video storyboards, websites, or mobile app UI/UX. Use when the user wants an original design language, identity, or art direction, or when a design feels generic, derivative, 'AI-default,' or inconsistent and needs one unifying idea. Triggers on 'design language', 'art direction', 'make it original', 'visual identity', 'looks generic', 'design a world'. Produces a completed Design Language Brief (core metaphor, visual vocabulary, design DNA, signature constraint, Originality Test verdicts) saved to the project and handed off to the production skill for the chosen medium.

Plan an animated product video end-to-end: pick the video type, define the animation style and pacing, write a beat-by-beat shot list, and produce ready-to-run keyframe prompts. Use whenever the user wants a product video, product animation, launch video, promo or ad video, animated product reveal, 360 spin, feature showcase, or 'a short video for my product' — even if they only say 'I need a video for the launch' or 'make my product look good in motion'. Also use to storyboard a video another tool will render. Produces a Product Video Production Spec: concept, beat sheet with timings and transitions, keyframe prompt list for fal-ai/nano-banana-pro, and audio/CTA direction.

Write a complete, render-ready Remotion video script as structured JSON: analyze the user's codebase or product, break the video into timed scenes, specify every visual element, animation, transition, and voiceover line, and list required assets and Remotion packages. Use whenever the user wants a Remotion video, a programmatic/code-rendered video, a product demo video of their app, an animated coding tutorial, a data-visualization video, or short-form animated social content — trigger on 'make a video of my app', 'Remotion', 'animated demo', 'code walkthrough video', 'render a video from code'. Produces a Remotion Script JSON (videoMetadata, scenes[], requiredAssets, technicalRequirements) that Remotion agent tooling can implement directly.

Produce platform-correct social media graphics: pick the right dimensions and layout per platform, write the headline/body/CTA copy, build the generation prompt, and generate the image via the repo's FAL.ai pipeline. Use whenever the user wants a social post visual, Instagram post/story/carousel, LinkedIn graphic, Twitter/X image, TikTok thumbnail, Pinterest pin, launch announcement graphic, or 'something to post about X' — even if they never name a platform. Also use to adapt one design across several platforms. Produces a Social Graphic Spec per asset (platform, dimensions, layout, copy, full generation prompt) plus generated images when FAL_API_KEY is available.

Plan presenter-led and UGC-style videos: choose the presenter archetype, define styling, setting, lighting, and framing, write the hook/body/CTA script skeleton, and produce keyframe prompts for the repo's FAL.ai pipeline. Use whenever the user wants a talking-head video, presenter video, UGC ad, testimonial, founder video, explainer with a person on camera, spokesperson content, or 'someone talking about my product' — even if they just say 'I need a video of a person explaining this'. Produces a Talking Head Video Spec: presenter definition, look/setting/lighting/framing decisions, timed script skeleton, and generation-ready keyframe prompts for fal-ai/nano-banana-pro.

Establishes a brand's visual direction and compresses it into a Creative Direction Brief with a paste-ready 'style block' that every downstream image prompt reuses, so all generated assets look like they came from one brand. Use when the user says 'define my brand style', 'I want a consistent look', 'what should my visuals look like', 'help me pick an aesthetic / mood / color palette', 'my generated images all look different', or before any multi-asset creative work where no style guide exists yet. Produces the Creative Direction Brief: aesthetic profile, palette with hex codes, style block for prompts, and prohibitions.

Turns any request for an AI-generated image into precise, generation-ready prompt specs and — when the FAL.ai automation is configured — the generated files themselves, using the single model fal-ai/nano-banana-pro. Use whenever the user asks to 'generate an image', 'make a picture of X', 'create a thumbnail / hero image / illustration / infographic', 'write me a prompt for this image', or any visual that isn't specifically product photography, a social-platform graphic, or a brand mark (hand those to the sibling skills). Produces one prompt-spec block per asset (subject, style, composition, lighting, mood, aspect ratio, negative constraints) plus file-naming and variant conventions, and runs docs/creative_cli.py to generate when a FAL_API_KEY is available.

Dispatcher for the creative suite: diagnoses a multi-asset creative request with a few questions, then routes it through the right sequence of creative skills, naming the artifact each step passes to the next, and wires in the FAL.ai automation (single model fal-ai/nano-banana-pro via docs/creative_cli.py) when available. Use when the user asks for several creative deliverables at once or an outcome that implies them — 'assets for my product launch', 'set up my brand visuals', 'I need a logo, product shots, and social posts', 'a week of content', 'rebrand everything' — or when it's unclear which creative skill applies. Produces a Creative Production Plan: chosen route, ordered skill sequence with hand-off artifacts, generation checklist, and asset output map. For a single obvious asset, skip this and use the specific skill directly.

Turn one piece of long-form content (blog post, video/podcast transcript, guide, talk) into a ready-to-post multi-platform content pack: LinkedIn posts, X/Twitter thread, newsletter item, short-video scripts, carousel outlines, quote pulls, and more — each natively adapted, not copy-pasted. Use when the user says 'repurpose this post', 'turn this into social content', 'atomize this', 'get more mileage out of my content', 'make LinkedIn posts from my article', or shares long-form content wanting distribution. Produces an idea inventory plus a per-platform atomization pack with a suggested posting order.

Write a complete automated email sequence, every email fully drafted: welcome sequence for new subscribers, ongoing nurture rotation, conversion sequence for an offer, or launch sequence for something new. Use when the user says 'write my welcome emails', 'set up a drip campaign', 'I'm launching X and need emails', 'nurture my list', 'write a sales email sequence', or connects a lead magnet to follow-up — even if they don't know which sequence type they need. Produces a sequence map (per-email purpose and send-day) plus full drafts: subject line, preview text, body, and one CTA per email.

Write a ready-to-send newsletter issue (and, when starting from zero, the recurring format and cadence to commit to): subject line options, preview text, body in one of six proven formats, and a single engagement CTA. Use when the user says 'write my newsletter', 'this week's issue', 'I want to start a newsletter', 'what should I send my list', 'help me email my subscribers regularly', or shares content to turn into an issue. Produces a complete issue draft with subject/preview variants — or a Newsletter Blueprint (format, cadence, section template) plus the first issue when setting one up.

Write a publish-ready, search-optimized article for a target keyword: SERP-validated brief, full draft matched to search intent, on-page metadata (title tag, meta description, URL slug), and schema markup. Use when the user says 'write a blog post about X', 'write an article that ranks', 'create SEO content', 'I need content for this keyword', or hands over a keyword/content brief from an SEO plan — even if they never say 'SEO'. Also use to rewrite or upgrade an existing article that isn't ranking. Produces the complete article plus a metadata block and an on-page checklist result.

Build a complete, prioritized SEO strategy for a site: topical authority map, winnable-keyword shortlist, content production pipeline, link-building plan, and technical baseline. Use when the user wants to grow organic traffic, asks 'how do I rank on Google', 'why isn't my site getting traffic', 'build me an SEO plan/strategy', 'what keywords should I target', 'how do I get backlinks', or wants to audit or scale their SEO — even if they never say the word 'strategy'. Produces an SEO Strategy Document with a scored keyword shortlist, a phased action plan sequenced by dependency, and a measurement plan.

Write high-engagement tweets and X/Twitter threads, grounded in live research of what's currently working in the user's niche: researches viral examples via web search, then drafts hook variants, a single-tweet version, and a thread version with alternates to test. Use when the user says 'write a tweet about X', 'turn this into a thread', 'help me go viral', 'grow my X/Twitter account', 'my tweets get no engagement', or shares an idea/article to tweet — any X/Twitter content request. Produces a Tweet Package: 3 hook options, final single tweet, full thread, and posting notes.

Define a brand's voice and produce a working voice guide: a personality archetype, settings on eight voice dimensions, do/don't rules, word choices, a tone matrix by context, and before/after rewrites of the user's real content. Use when the user says 'define my brand voice', 'what should our tone be', 'make this sound like us', 'our content sounds generic/inconsistent', 'write tone-of-voice or brand guidelines', or before producing copy at scale across channels. Also use in audit mode when existing content needs to be checked against a voice. Produces a Voice Guide document that other skills consume when writing any copy.

Write copy engineered to drive one specific action — landing pages, sales pages, opt-in pages, ads, promotional emails — using an eight-part direct-response architecture (hook, problem, mechanism, solution, proof, objections, CTA, urgency). Use when the user says 'write my landing page / sales page / ad', 'this page isn't converting', 'rewrite my homepage copy', 'write copy for my lead magnet page', or any request for persuasive copy with a measurable action at the end. Produces the full copy draft, section-labeled against the architecture, plus hook alternatives and a conversion checklist.

Build a prioritized keyword cluster map using the 6 Circles Method — keywords at the intersection of positioning alignment, customer language, competitor blind spots, workable search volume, real expertise, and business intent. Use when the user asks 'what keywords should I target', 'what should I write about', 'find SEO opportunities', 'plan my content strategy', 'what does my audience search for', or before any SEO content is written. Produces a Keyword Map: scored keywords grouped into topic clusters, each cluster mapped to the article that should target it, ready to hand to content production.

Design a lead magnet — the free offer that trades real value for an email address — by picking the right problem slice, hook, and format, validating it, and drafting the asset. Use when the user says 'grow my email list', 'I need a lead magnet / freebie / opt-in', 'create a free download', 'how do I capture leads', 'nobody signs up', or when a funnel needs an entry offer before the paid ask. Produces a Lead Magnet Spec (title, hook, format, contents outline, delivery, follow-up path) plus a draft of the asset itself when the format allows.

Diagnose a marketing situation and route it to the right marketing skill sequence. Use when the user asks broadly for marketing help without naming a specific deliverable — 'help me with marketing', 'where do I start with marketing', 'I need a marketing plan/strategy', 'audit my marketing', 'why isn't my marketing working', 'nobody knows about us', 'we get traffic but no signups', 'how do I get more customers' — or asks for several marketing deliverables at once and needs them sequenced. Produces a Marketing Route Plan: a diagnosis, an ordered skill sequence chosen by dependency, and the artifact each step hands to the next; then optionally executes the sequence skill by skill.

Find and state a market positioning that differentiates the user's offer: match the message to the market's maturity stage, name the unique mechanism, choose one psychological angle, and pressure-test the result. Use when the user asks 'how do I position this', 'how do I stand out', 'why should anyone pick us', 'write my value prop / tagline / headline claim', 'our marketing sounds like everyone else's', or when any copy or content work stalls because there's no clear core claim — even if they never say the word positioning. Also works in audit mode on an existing positioning. Produces a Positioning Statement: market stage, named mechanism, chosen angle with the claim written out, and a pass/fail test record.

Evaluate, pressure-test, or sharpen any product, feature, app, startup, or business idea using the Foundation Sprint method from Jake Knapp and John Zeratsky's book Click (How to Make What People Want). Use this whenever the user wants to decide what to build, judge whether an idea is worth pursuing, find product-market fit, nail differentiation, choose between approaches or directions, write or critique a value proposition, or make an existing product better. Trigger on phrases like 'is this a good idea', 'should I build this', 'validate this idea', 'what should I build', 'how do I stand out', 'how do I differentiate', 'will people want this', 'evaluate my idea', 'critique my product', 'how do I improve my product', 'founding hypothesis', 'positioning', or any time someone is starting a project or stuck choosing a direction, even if they never mention the book. Produces a testable Founding Hypothesis, a differentiation 2x2, a Magic Lenses comparison, and a prove-it scorecard.