Communitygithub.com

NeverSight/learn-skills.dev

Static artifact craft skill for self-contained HTML/CSS/JS documents: docs, sheets, dashboards, explainers, slides, tools, and landing pages. Use when the user asks for a durable, openable, shareable web deliverable they'll keep or hand off — a report, a dashboard, a slide deck, a data table, a page. Local folder first, temporary public link via tunnel (localhost.run), optional durable publish to Surge, GitHub Pages, or Cloudflare. Not for quick look renders, inline snippets, or throwaway scratch. Not for SPA frameworks, backend APIs, database apps, or production product UI.

What is learn-skills.dev?

learn-skills.dev is a Claude Code agent skill that static artifact craft skill for self-contained HTML/CSS/JS documents: docs, sheets, dashboards, explainers, slides, tools, and landing pages. Use when the user asks for a durable, openable, shareable web deliverable they'll keep or hand off — a report, a dashboard, a slide deck, a data table, a page. Local folder first, temporary public link via tunnel (localhost.run), optional durable publish to Surge, GitHub Pages, or Cloudflare. Not for quick look renders, inline snippets, or throwaway scratch. Not for SPA frameworks, backend APIs, database apps, or production product UI.

Works with✓Claude Code~Codex CLI~Cursor
npx skills add https://github.com/NeverSight/learn-skills.dev/tree/HEAD/data/skills-md/1broseidon/skills/folio

Ask in your favorite AI

Open a new chat with this agent skill pre-loaded.

Documentation

Folio

Read the request and any existing artifact before writing HTML. Every external dependency and publish URL belongs in the manifest ledger. Artifacts are static documents — they do not call APIs.

When to use folio — and when not to

Folio is for durable, openable, shareable web deliverables — things the user will keep, reopen, or hand to someone else.

Use folio when:

  • The user asks for a doc, dashboard, sheet, deck, or page they'll save or send
  • The output is a deliverable, not a quick look ("make me a report", "build a dashboard for the team", "put this online", "I want a page that…")
  • The user invokes a verb explicitly (folio publish, folio audit, folio remix, folio describe)

Do not use folio when:

  • The user wants a quick look, not a deliverable ("show me", "what does this look like", "just check", "can you display")
  • The output won't be opened again after this conversation — answer inline or write a scratch file, don't stamp it, ledger it, or gate it
  • The request is for a code snippet, a config file, a single function — those aren't browser artifacts

Entrance gate. Before entering the artifact flow, ask: will the user open this file again after this session? If no, folio is the wrong tool — just answer. If yes, proceed.

Standalone is the default. The lightest folio path is a single folder with one index.html — no site root, no site.json, no index page. A site root appears only when there's a second artifact or the user asks to publish a collection. Do not create site ceremony for one file.

Glossary

Use only these terms in Folio output.

TermMeaning
artifactA self-contained static web deliverable: HTML with CSS/JS, runnable in a browser
bundleOne of single-file (inline CSS/JS) or multi-file (linked assets in one directory)
artifact kindOne of doc, sheet, dashboard, explainer, slides, tool, landing
host profileWhere the site lives durably: local, surge, gh-pages, cf-pages, cf-worker
share (tunnel)A temporary public URL to the running local server, via localhost.run (default) or ngrok. Ephemeral, machine stays host — not a publish
site rootA directory containing an index.html and an artifacts/ tree, published as one domain. Artifacts are paths under it, not separate subdomains.
site manifest.folio/site.json — the canonical source of truth. Carries domain + per-artifact registry (slug, kind, bundle, ledger, provenance, publish state). Stamps, index, and audit all derive from it.
standalone artifactAn artifact with no site root. The HTML stamp is its only manifest. Upgrades to a site entry when a second artifact appears.
provenanceThe source field on a remixed artifact: { slug, site, url }. Null for originals. Read by describe, audit, and remix.
manifestEntry HTML, file list, external deps, and optional publish metadata
ledgerThe manifest ledger — deps, data files, URLs, and evidence levels before build
gateA binary check that must pass before handoff
evidence levelOne of observed, derived, stated, absent

How to use this skill

  • Default: create or update one artifact. Follow the artifact flow. A single artifact is standalone by default — no site root required.
  • folio audit <path|url>: read-only audit. Do not edit. Read references/verbs/audit.md.
  • folio describe <path|url|slug>: read an artifact's manifest entry (kind, ledger, provenance, publish state). Read references/verbs/describe.md.
  • folio share <site-root>: open a temporary public link to the locally running site via a tunnel (default localhost.run — no install, no account; --ngrok alt). The machine stays the host; the URL dies with the process. Use this for "let someone see it right now," not publish. Read references/verbs/share.md.
  • folio publish <site-root>: deploy a site root (all artifacts) to the chosen host profile for durable hosting elsewhere. Regenerate the index first (scripts/gen-index.js), then deploy. Read references/verbs/publish.md.
  • folio remix <path|url>: fork an artifact and change it while preserving manifest honesty and recording source lineage. Read references/verbs/remix.md.

One run = one artifact unless the user names multiple. Prefer the smallest bundle that satisfies the request. A single artifact is standalone; multiple artifacts live under one site root and publish as one domain. Read references/site-index.md for the site-root model and standalone-to-site upgrade.


Non-negotiable disciplines

  1. Runnable before pretty. The artifact opens in a browser and core interactions work before polish or publish.
  2. Static documents only. No fetch() to external APIs, no WebSocket connections, no XMLHttpRequest to live endpoints. Data is embedded inline, loaded from a local file via fetch('./data/…'), or generated at build time. The only network requests an artifact makes are to ledgered CDN scripts, styles, and fonts.
  3. Manifest honesty. CDN scripts, fonts, images, and data files are ledger rows, not surprises.
  4. Local-first, then share, then publish. Default host profile is local. Run it the dynamic way — node <folio-skill>/scripts/serve.js <site-root> ensures the artifacts/ folder, regenerates the index/search page from the ledger, picks an open port, and serves over HTTP (plain file:// is fine for a fully inlined single file, but breaks fetch and module imports). See references/local-server.md. To show someone right now, open a temporary tunnel — node <folio-skill>/scripts/share.js <site-root> (default localhost.run, no install/account); the link is ephemeral and the machine stays the host. Durable hosting elsewhere is publish — only when the user asks or folio publish is invoked. Never advertise share or publish capabilities proactively.
  5. No secrets in static files. API keys, tokens, and private URLs do not belong in HTML/JS committed or published.
  6. Not ugly. Artifacts follow the visual baseline. Read references/visual-baseline.md before build.

Gold decision trace

Request: Build an interactive latency histogram from this CSV for a teammate.

Inventory:
· Target: new artifact `latency-hist`
· Existing: none
· Data: user attached `runs.csv` (observed, 840 rows)
· Constraints: local folder, no publish

Picks: kind=explainer · bundle=multi-file · host=local · audience=teammate handoff

Ledger:
dep | kind | evidence | in artifact as
d3@7 (cdn.jsdelivr) | script | stated, user did not pin | index.html script tag + note in README
runs.csv | data | observed | ./data/runs.csv, loaded via fetch('./data/runs.csv')

File plan:
Create: artifacts/latency-hist/index.html, style.css, app.js, data/runs.csv, README.md
No deletes.

Verification: npx serve artifacts/latency-hist; chart renders; console clean
Gates: opens-locally=yes, console-clean=yes, ledger-complete=yes, no-secrets=yes, mock-labeled=n/a, visual-baseline=yes

Handoff: path, how to open, ledger summary, deferred: none

Scope fork

Resolve scope before broad design questions.

Request shapeBundleHostSkip
Quick throwaway demosingle-filelocalREADME, publish, multi-page
Written doc or reportsingle-file or multi-filelocalCharts, heavy JS
Data table or sheetsingle-file or multi-filelocalNarrative prose, decoration
Metrics dashboardmulti-filelocalFull CRUD, live data feeds
Concept explainermulti-filelocalDense raw data dumps
Slide deckmulti-filelocalScroll layouts, widgets
Shared teammate artifactmulti-filelocalFramework build step
Show someone right nowsingle-file or multi-fileshare (tunnel)Deploy ceremony, durable host
Durable public linkmulti-filesurge or cf-pagesAuth, server logic
Agent handoff previewsingle-file preferredlocalShare/publish unless asked

Ambiguous request: ask once, tersely. Example: "Single self-contained file or a small folder with separate CSS/JS?" If unanswered, choose the smallest bundle that works.


Artifact flow

0. Inventory

Read the request and any existing artifact tree. Read references/evidence.md and references/artifact-types.md.

Collect:

  • Subject: what the artifact should show or let the user do
  • Inputs: files, JSON, CSV, copy, design constraints (with evidence levels)
  • Existing artifact: path, entry HTML, manifest, last publish URL if any
  • Host profile: default local; read references/host-profiles.md when publish is in scope
  • Convention source: folio.md if present

Emit a compact inventory with citations before proceeding.

1. Picks

Emit before building: Picks: kind=<x> · bundle=<y> · host=<z> · audience=<a>.

Read references/artifact-types.md when kind is unclear. Read references/host-profiles.md when host is not local.

2. Manifest ledger

Build the ledger before writing HTML. Read references/manifest-ledger.md.

Columns:

dep | kind | evidence | in artifact as

Kinds: script, style, font, image, data, mock, publish-target.

Data files and mock data must be explicit. External API calls are out of scope — see discipline #2.

3. File plan and build

State the file plan before editing:

Create: artifacts/<slug>/index.html, app.js, data/sample.json
Modify: none
No deletes.

Start from a template. Load references/templates/<kind>.html and adapt it — replace the {{placeholder}} content and <!-- Replace: ... --> markers with real content. Templates are gate-passing folio artifacts themselves; they carry the token block, the light/dark toggle, the favicon, and the kind-specific structure. Read references/templates/ index for what each template provides.

Read references/visual-baseline.md before writing CSS.

Build with plain HTML/CSS/JS unless the user named a specific stack. Prefer:

  • Semantic HTML, one obvious entry file (index.html)
  • CSS custom properties for all visual tokens (color, type, spacing, radius, shadow)
  • OKLCH for color; no inline hex/rgb outside the :root token block
  • Vanilla JS or a single named CDN library — ledgered
  • multi-file: keep all asset paths relative; no absolute machine paths
  • No italic headings; readable measure (max-width: 60–70ch); overflow-x: clip on html and body

Read references/security.md when the artifact accepts user input, embeds third-party content, or publishes publicly.

Stamps (HTML comment, top of entry file):

<!-- Folio · slug: latency-hist · kind: explainer · bundle: multi-file · host: local -->
<!-- deps: d3@7 · data: runs.csv · publish: none -->

For single-file, keep the stamp minimal (one line).

4. Verification

Read references/verification.md. Run checks that fit the host profile.

Minimum matrix:

check | command/evidence | result | notes

Examples: static server + load entry URL, DevTools console clean, click through primary interaction, broken link scan, file:// vs server note for fetch, visual-baseline spot-check.

5. Gates

Apply before handoff. Any no triggers a fix loop.

GateCheck
opens-locallyEntry HTML loads via static server (or honest file:// caveat documented); for server setup see references/local-server.md (scripts/serve.js is the one-command path)
console-cleanNo uncaught errors on primary path
ledger-completeEvery external dep and data source appears in the ledger
no-secretsNo keys, tokens, or private URLs in source
mock-labeledMock/sample data labeled when real data is absent
static-onlyNo fetch() to external APIs, no WebSocket, no XHR to live endpoints
visual-baselineToken block present, no inline colors outside :root, no italic headings, contrast adequate
publish-readyWhen publishing: host profile steps documented and verified or explicitly deferred
manifest-derivationAdvisory. HTML stamp (slug, kind, bundle) matches the site.json entry for this artifact. Flags drift; see references/verification.md#manifest-derivation-check.

Read references/slop-test.md after building. Fix P0/P1 before handoff.

6. Publish (optional)

Only when the user requests publish or invokes folio publish. Read references/verbs/publish.md and the host profile deep file.

Record publish URL in the manifest and handoff. Never claim a URL without evidence from the publish command output.

7. Handoff

Include:

  • Artifact path and entry file
  • How to open locally (exact command)
  • Publish URL (if published), with evidence
  • Manifest ledger summary
  • Verification commands and results
  • Deferred items (e.g., fetch requires server, mock data in use)

Do not bury manifest or publish facts under implementation chatter.


BAD/GOOD contrasts

BAD: Wire fetch('/api/users') because dashboards usually load users from an API. GOOD: Embed data/users.json with a visible "Sample data" label. No API calls.

BAD: Paste Google Analytics, fonts, and three CDN libs without listing them. GOOD: Each dep is a ledger row with URL and purpose.

BAD: "Deployed at https://demo.example.com" with no publish command run. GOOD: Publish URL copied from command output, or marked deferred.

BAD: Inline oklch() values scattered through the CSS, three different font-family declarations. GOOD: All visual values in the :root token block, referenced via var().


Convention artifacts

folio.md (output dir, default host, CDN allowlist, naming): opt-in. Read references/folio-md.md. Once present, house defaults win.

Optional per-project layout:

site-root/
  index.html          # generated by scripts/gen-index.js (filter + kind groups)
  artifacts/
    <slug>/
      index.html
      ...
.folio/
  site.json           # site manifest: domain + artifact registry

Slug convention: <kind>-<descriptor> (e.g., explainer-folio-loop) or <descriptor> alone when the kind is obvious (e.g., chart-297). The kind always lives in the stamp and site manifest.


Reference loading rules

Keep SKILL.md lean. Load before proceeding when a path is listed.

Always load: references/evidence.md, references/manifest-ledger.md, references/verification.md, references/visual-baseline.md.

Load when applies:

  • Kind choice: references/artifact-types.md
  • Build: references/templates/<kind>.html — the gate-passing starter for the chosen kind
  • Local run / serving: references/local-server.md (run modes, scripts/serve.js, runtime probe)
  • Share / temporary link: references/verbs/share.md (tunnel via scripts/share.js, localhost.run default, ngrok alt)
  • Host/publish: references/host-profiles.md, references/verbs/publish.md, references/site-index.md
  • User input, share, or public publish: references/security.md
  • Verbs: references/verbs/audit.md, references/verbs/describe.md, references/verbs/remix.md
  • Anti-patterns: references/anti-patterns.md (before build), references/slop-test.md (after build)
  • Conventions: references/folio-md.md

Never load at runtime: references/philosophy.md, references/examples/ (human-only worked examples), examples/ (human-only).


v2 limits

In scope: static HTML/CSS/JS artifacts (doc, sheet, dashboard, explainer, slides, tool, landing), single-file and multi-file bundles, local folder output, temporary public link via tunnel (localhost.run default, ngrok alt), optional durable Surge / GitHub Pages / Cloudflare Pages publish, audit/describe/share/remix/publish verbs, manifest ledger, vanilla JS + ledgered CDN libs, visual baseline (OKLCH tokens, system fonts, anti-AI-slop rules).

Deferred: external API integration (v0.3 opt-in with evidence + offline degradation), React/Vue/Svelte build pipelines, SSR, databases, WebSocket backends, artifact versioning UI, collaborative editing, automated screenshot regression.

Out of scope: production app UI, backend APIs, ORM/data layers, native mobile, PDF/Office artifacts (see other skills).

Individual skills in this repo

This repo contains 20 individual skills — each has its own dedicated page.

NeverSight/learn-skills.dev

Use when generating or modifying Remotion video code, creating demo videos, or working with the demo-video/ directory

NeverSight/learn-skills.dev

Create AI avatar and talking head videos via inference.sh CLI. Recommended: P-Video-Avatar (fastest, cheapest, built-in TTS). Also: OmniHuman, Fabric, PixVerse. Audio: Inworld TTS-2 (100+ languages, emotion steering for characters), ElevenLabs, Kokoro. Capabilities: audio-driven avatars, text-to-avatar, lipsync videos, talking head generation, virtual presenters, UGC content. Use for: AI presenters, explainer videos, virtual influencers, dubbing, marketing videos, UGC ads, gaming avatars, NPC dialogue. Triggers: ai avatar, talking head, lipsync, avatar video, virtual presenter, ai spokesperson, audio driven video, heygen alternative, synthesia alternative, talking avatar, lip sync, video avatar, ai presenter, digital human, ugc, ugc video, ugc ad, avatar ugc

NeverSight/learn-skills.dev

Create AI marketing videos for ads, promos, product launches, and brand content. Models: Veo, Seedance, Wan, FLUX for visuals, Kokoro for voiceover. Types: product demos, testimonials, explainers, social ads, brand videos. Use for: Facebook ads, YouTube ads, product launches, brand awareness. Triggers: marketing video, ad video, promo video, commercial, brand video, product video, explainer video, ad creative, video ad, facebook ad video, youtube ad, instagram ad, tiktok ad, promotional video, launch video

NeverSight/learn-skills.dev

Generate AI videos with Google Veo, Seedance 2.0, HappyHorse, Wan, Grok and 40+ models via inference.sh CLI. Models: Veo 3.1, Veo 3, Seedance 2.0, HappyHorse 1.0, Wan 2.5, Grok Imagine Video, OmniHuman, Fabric, HunyuanVideo. Capabilities: text-to-video, image-to-video, reference-to-video, video editing, lipsync, avatar animation, video upscaling, foley sound. Use for: social media videos, marketing content, explainer videos, product demos, AI avatars. Triggers: video generation, ai video, text to video, image to video, veo, animate image, video from image, ai animation, video generator, generate video, t2v, i2v, ai video maker, create video with ai, runway alternative, pika alternative, sora alternative, kling alternative, seedance, happyhorse

NeverSight/learn-skills.dev

ElevenLabs automatic dubbing - translate and dub audio/video into 29 languages while preserving speaker voice via inference.sh CLI. Capabilities: auto speaker detection, voice-preserving translation, video dubbing, audio localization. Use for: content localization, video translation, multilingual content, international distribution. Triggers: dubbing, dub video, translate audio, video translation, audio translation, localize content, elevenlabs dubbing, eleven labs dub, multilingual dub, voice translation, auto dub, language dub, content localization

NeverSight/learn-skills.dev

Explainer video production guide: scripting, voiceover, visuals, and assembly. Covers script formulas, pacing rules, scene planning, and multi-tool pipelines. Use for: product demos, how-it-works videos, onboarding videos, social explainers. Triggers: explainer video, how to make explainer, product video, demo video, video production, video script, animated explainer, product demo video, tutorial video, onboarding video, walkthrough video, video pipeline

NeverSight/learn-skills.dev

Still-to-video conversion guide: model selection, motion prompting, and camera movement. Covers Wan 2.5 i2v, Seedance, Fabric, Grok Video with when to use each. Use for: animating images, creating video from stills, adding motion, product animations. Triggers: image to video, i2v, animate image, still to video, add motion to image, image animation, photo to video, animate still, wan i2v, image2video, bring image to life, animate photo, motion from image

NeverSight/learn-skills.dev

Generate talking head avatar videos with Pruna P-Video-Avatar via inference.sh CLI. Turn a portrait image into a realistic speaking video with built-in TTS. 18x faster and 6x cheaper than competitors. Models: P-Video-Avatar, P-Image (for portrait generation). Capabilities: text-to-avatar, audio-driven avatars, 30 voices, 10 languages, 720p/1080p, built-in TTS, dynamic backgrounds, full-body control. Use for: AI presenters, product demos, explainer videos, virtual influencers, marketing, education, multilingual content, UGC, gaming avatars. Triggers: avatar video, talking head, ai avatar, p-video-avatar, pruna avatar, video avatar, ai presenter, digital human, virtual presenter, lipsync, talking avatar, ai spokesperson, heygen alternative, synthesia alternative, veed alternative, fabric alternative, omnihuman alternative

NeverSight/learn-skills.dev

Generate videos with Pruna P-Video and WAN models via inference.sh CLI. Models: P-Video, WAN-T2V, WAN-I2V. Capabilities: text-to-video, image-to-video, audio support, 720p/1080p, fast inference. Pruna optimizes models for speed without quality loss. Triggers: pruna video, p-video, pruna ai video, fast video generation, optimized video, wan t2v, wan i2v, economic video generation, cheap video generation, pruna text to video, pruna image to video

NeverSight/learn-skills.dev

Render videos from React/Remotion component code via inference.sh. Pass TSX code, get MP4. Supports all Remotion APIs: useCurrentFrame, useVideoConfig, spring, interpolate, AbsoluteFill, Sequence. Configurable resolution, FPS, duration, codec. Use for: programmatic video generation, animated graphics, motion design, data-driven videos, React animations to video. Triggers: remotion, render video from code, tsx to video, react video, programmatic video, remotion render, code to video, animated video, motion graphics code, react animation video

NeverSight/learn-skills.dev

Video ad creation with exact platform-specific specs for TikTok, Instagram, YouTube, Facebook, LinkedIn. Covers dimensions, duration limits, AIDA framework, and caption requirements. Use for: video ads, social media ads, paid media creative, video marketing, ad production. Triggers: video ad, social media ad, tiktok ad, instagram ad, youtube ad, facebook ad, linkedin ad, video creative, ad specs, paid media, video marketing, ad production, reels ad, stories ad, pre roll, bumper ad

NeverSight/learn-skills.dev

Best practices and techniques for writing effective AI video generation prompts. Covers: Veo, Seedance, Wan, Grok, Kling, Runway, Pika, Sora prompting strategies. Learn: shot types, camera movements, lighting, pacing, style keywords, negative prompts. Use for: improving video quality, getting consistent results, professional video prompts. Triggers: video prompt, how to prompt video, veo prompts, video generation tips, better ai video, video prompt engineering, video prompt guide, video prompt template, ai video tips, video prompt best practices, video prompt examples, cinematography prompts

NeverSight/learn-skills.dev

YouTube thumbnail design with specific dimensions, contrast rules, and mobile preview optimization. Covers safe zones, text placement, face expression psychology, and A/B testing. Use for: YouTube thumbnails, video cover images, click-through optimization. Triggers: youtube thumbnail, thumbnail design, video thumbnail, click through rate, ctr optimization, youtube cover, video cover image, thumbnail maker, thumbnail tips, youtube design, video preview image

NeverSight/learn-skills.dev

Landing page conversion optimization with layout rules, hero section design, and CTA psychology. Covers above-the-fold formula, social proof placement, mobile design, and F-pattern reading. Use for: startup landing pages, product pages, SaaS marketing, conversion optimization. Triggers: landing page, hero section, above the fold, conversion optimization, landing page design, cta button, hero image, landing page layout, saas landing page, product page design, conversion rate, landing page best practices

NeverSight/learn-skills.dev

Configure and use the hosted YouTube Data MCP end-to-end with minimal user input. Use when users want the agent to verify Node.js and `npx`, configure MCP server config (Windows/macOS, Cursor/Codex/OpenClaw/OpenCode), request API key at setup time, run post-install capability discovery (`tools/list` and `get_patch_notes`), and then strongly recommend helper skill and Python setup for full local document and spreadsheet workflows.

NeverSight/learn-skills.dev

Creates 120fps GPU-accelerated animations with Motion.dev (Framer Motion successor) for React, Next.js, Svelte, and Astro projects. Use when user requests animation, motion, scroll effects, parallax, hero animations, gestures, drag interactions, spring physics, whileHover effects, whileInView animations, animated UI, micro-interactions, page transitions, or layout animations. Generates production TypeScript/JSX code with accessibility (prefers-reduced-motion) and performance validation (≥60fps). Supports entrance animations, gesture interactions (hover/tap/drag), scroll-based reveals, and layout transitions using spring physics and natural timing. Do NOT use for CSS-only transitions (use native CSS), static sites without JavaScript, Vue animations (use motion-v variant instead), or SVG/Canvas complex animations (GSAP better suited).

NeverSight/learn-skills.dev

FFmpeg commands for video/audio conversion, trimming, compression, and processing. Use when user mentions "ffmpeg", "convert video", "compress video", "extract audio", "trim video", "gif from video", "video codec", "transcode", "screen recording", "merge videos", "video to mp4", "reduce file size", or any media processing task.

NeverSight/learn-skills.dev

Vim keybindings, motions, text objects, and operators for efficient text editing. Use when user asks about "vim commands", "vim motions", "text objects", "vim keybindings", "vim cheat sheet", "learn vim", "vim in VS Code", or any Vim editing tasks.

NeverSight/learn-skills.dev

Download YouTube videos, extract/proofread/translate subtitles, and render them onto video. Use whenever the user asks to download a YouTube video with subtitles, translate video subtitles to Chinese, add subtitles/burn subtitles to a video, or do ASR transcription on video audio. Covers the full pipeline: video download → subtitle extraction → proofreading → translation → SRT generation → subtitle rendering. Also use for "下载视频加字幕", "视频翻译字幕", "把字幕烧录到视频中". For pure subtitle file creation without rendering (just SRT output), this skill handles that too — stop before Phase 5.

NeverSight/learn-skills.dev

Grok Build ONLY. Turn a 2D character still into smooth animation sprites via image_gen/image_edit base → image_to_video (6s/10s run-in-place) → ffmpeg frames → magenta chroma-key → dense sampled sprites (strip/grid/GIF). Use when the user wants video-to-sprite, motion capture from generated video, smoother run/walk cycles from dense frames, or runs /video2dsprite. Do NOT use on Codex/Claude — only Grok Build has image_to_video. Prefer generate2dsprite for crisp pixel sheets without video.

Related Skills