@heygen-com

tailwind covers browser-runtime tailwind v4 styling for hyperframes compositions, with explicit version pinning, css-first theme tokens, dynamic class safety guardrails, and deterministic render validation.

slideshow enables authoring discrete, navigable presentation decks within HyperFrames compositions using a JSON island schema for slide declaration, fragments, branching hotspots, and presenter notes. supports multi-branch narratives, fragment-based reveals, and synchronized dual-window presenter mode.

three covers deterministic three.js scene rendering within hyperframes compositions via seek-event synchronization. establishes a contract for time-based rendering, animation mixer patterns, and asset loading constraints.

remotion-to-hyperframes converts react-based remotion video compositions to hyperframes html with frame-accuracy validation. translates ~80% of compositions mechanically, detects untranslatable patterns, and refuses output without running test suite.

contribute-catalog guides users through authoring and shipping new hyperframes registry blocks or components, from initial clarification through peer review via pull request. covers scaffolding, build templates, validation tooling, and merge requirements.

hyperframes-cli scaffolds video compositions with init templates, lints for structural errors, previews with hot-reload, and renders to multiple formats. environment checks and audio tasks delegate to hyperframes-media.

seam-craft defines render prerequisites and master-timeline mechanics for scene-to-scene transitions in HyperFrames, covering opacity guards, wrapper overlap scheduling, track ping-pong assignment, and gsap template injection with token substitution.

website-to-hyperframes converts urls into branded video content through a 7-step workflow combining storyboarding, voice generation, and composition validation, with toggleable collaborative vs autonomous modes.

hyperframes-read-first routes video creation requests to appropriate sub-skills via a capability map and documents the composition authoring contract, but lacks step-by-step procedures and explicit failure handling.

motion-graphics routes short, design-led motion graphics under 10 seconds where motion carries meaning. it dispatches to specialized video skills for narrated, multi-scene, or interactive content, and delegates asset sourcing and composition to subagents.

waapi enables synchronous keyframe animation control via web animations api seeking, integrated with hyperframes timeline positioning and deterministic state persistence.

graphic-overlays layers timed, designed graphic cards (titles, lower-thirds, callouts, side panels) onto existing video, synced to transcript. the agent designs each card in html and renders the composition to mp4 via hyperframes.

music-to-video orchestrates beat-synced video creation from audio tracks, analyzing rhythm data once then dispatching frame composition work to sub-agents. input is music plus optional user media, output is a beat-gridded html5 video.

embedded-captions applies caption styles (standard, cinematic, themed) to talking-head video via deterministic pipeline. selects from 36 identity presets, composes typography into scene, and outputs final composite without touching original footage.

video-download wraps yt-dlp to fetch videos, audio, and subtitles from urls at configurable resolutions. provides command templates for common tasks but lacks structured intent/input/output contracts.

hyperframes-registry installs reusable blocks and components into HyperFrames compositions via cli. blocks require wiring attributes; components need manual html, css, javascript merging into host compositions.

figma imports design assets, tokens, components, and motion data into hyperframes compositions through phased REST transport, with optional connector support for advanced motion and shader handling.

captions-overlay establishes a doctrine for layering captions as composited overlays atop video content without reserving screen real estate. it defines three caption modes (drop, rail, embed) with strict spacing rules and forbids layout shifts to accommodate text.

css-animations provides deterministic seeking and preview support for keyframe animations in HyperFrames, suited to decorative single-element motion with fixed duration and finite iteration counts.

typegpu enables gpu-accelerated compositions in hyperframes via webgpu or typegpu, requiring manual event listening and explicit synchronization between timeline and render calls.

hyperframes-media provides three local cli commands for media preprocessing: text-to-speech with 54 multilingual voices, audio transcription with word-level timestamps, and video background removal for transparent overlays. no api keys required.

claude-design-hyperframes generates valid html/css/gsap video drafts for claude design contexts, positioning itself as a rapid brand-aware prototyping layer that hands off to claude code for animation refinement.

heygen-avatar creates and manages digital avatars for agents, users, or named characters, persisting identity data to markdown files for reuse across sessions.

avatar-video wraps heygen api endpoints for creating animated videos from avatars or images, supporting voice synthesis and script control via two discriminated union types.

faceless-explainer converts arbitrary text into faceless video with llm-generated visuals and auto-selected style presets. includes routing logic to redirect product/site/code content to specialized skills.

hyperframes-core defines the html composition contract for seekable video rendering via dom data attributes and framework-owned media playback, with animation runtime and timing primitives delegated to sibling skills.

motion-doctrine establishes continuity rules for multi-scene video by prescribing transition choices and entry/exit mechanics at scene seams. gateway skill that routes to implementation tools rather than prescribing direct actions.

product-launch-video routes promotional video requests to a hyperframes orchestrator, capturing product details and constructing frame-by-frame output, but the procedure is incomplete and assumes deep familiarity with undocumented reference files.

create-video delegates video generation to heygens api via text prompt. covers authentication setup and tool preference but lacks procedural depth, failure handling, and concrete usage examples.

visual-style defines portable design systems as markdown. it maps colors, typography, layout, motion and mood into a consumable file format without explaining how those systems actually integrate or execute.

cut-the-curve catalogs seven motion-transition techniques (zoom-through, inverse zoom, curve cuts, waterfall, rack-focus, entry, nudge) with axes and use cases, but omits procedural steps, inputs, and decision logic.

media-use orchestrates asset resolution across bgm, sfx, image, icon, voice, and avatar types through heygen cli binding. scope unclear, procedure incomplete, edge cases absent.

An orbiting camera follows an ice logo as it breaks apart, reforms into two text moments, and fades. HyperFrames block, 1920×1080, 22.5s, 216 variables.

A single feature card orbits a dot sphere on approved Blender camera motion. HyperFrames block, 1920×1080, 10s, 4 variables.

A wireframe portal bursts open, the title comes through, then its letters swap into a second phrase. HyperFrames block, 1920×1080, 8s, 9 variables.

Leaves sweep through the frame and part to reveal the headline. HyperFrames block, 1920×1080, 12s, 11 variables.

A tiled headline surface flips cell by cell under a sweeping depth field to reveal the rear headline. HyperFrames block, 1920×1080, 8s, 18 variables.

Glass shards fly in through fog and tile themselves into the headline. HyperFrames block, 1920×1080, 12.16s, 19 variables.

A chain of bevelled cuboids rides a travelling wave as a content carousel. HyperFrames block, 1920×1080, 6.666666666666667s, 40 variables.

oversized-cursor — an installable skill for AI agents, published by heygen-com/hyperframes.

changelog-video — an installable skill for AI agents, published by heygen-com/hyperframes.

Lottie and dotLottie adapter patterns for HyperFrames. Use when embedding lottie-web JSON animations, .lottie files, @lottiefiles/dotlottie-web players,…

Anime.js adapter patterns for HyperFrames. Use when writing Anime.js animations or timelines inside HyperFrames compositions, registering animations on…

GSAP animation reference for HyperFrames. Covers gsap.to(), from(), fromTo(), easing, stagger, defaults, timelines (gsap.timeline(), position parameter,…

Capture a general website/URL and turn it into a video OF the site — tour, showcase, or social clip built from captured screenshots and the site's own brand…

Use when working with a person on a HyperFrames project in Studio: first, whether their message asks for a change at all (questions, loose ideas and "don't…

Package an existing talking-head / interview / podcast video with timed, designed GRAPHIC OVERLAY cards — kinetic titles, lower-thirds, data callouts, quotes,…

Turn a GitHub pull request (a PR URL, owner/repo#N, or 'this PR' in a checked-out repo) into a code-change explainer video — changelog, feature reveal, fix, or…

Use when audio already placed in a HyperFrames composition needs to be mixed: fade-in/fade-out, crossfade, track gain or volume, volume automation, ducking, a…

Author or edit a custom HyperFrames composition when no specialized workflow fits, or when BRIEF.md sets flow: companion. Use for longer or multi-scene pieces,…

Use when a HyperFrames composition needs a punch-in, punch-out, zoom, reframe, Ken Burns treatment, camera move, visual match/whip handoff, or other seek-safe…

Non-animation creative direction for HyperFrames videos. Use for design spec (frame.md / design.md) handling, palettes, typography, narration, beat planning,…

All animation knowledge for HyperFrames — atomic motion rules, multi-phase scene blueprints, scene transitions, broader motion-design techniques, AND the seven…

Mandatory entry point: read this first for any request to make, create, edit, animate, or render a video, animation, or motion graphic, including a promo,…

[DEPRECATED] Use `create-video` for prompt-based video generation or `avatar-video` for precise avatar/scene control. This legacy skill combines both workflows…

faceswap — an installable skill for AI agents, published by heygen-com/skills.

ai-video-gen — an installable skill for AI agents, published by heygen-com/skills.

video-edit — an installable skill for AI agents, published by heygen-com/skills.

video-understand — an installable skill for AI agents, published by heygen-com/skills.

video-translate — an installable skill for AI agents, published by heygen-com/skills.

text-to-speech — an installable skill for AI agents, published by heygen-com/skills.

heygen-skills — an installable skill for AI agents, published by heygen-com/skills.

heygen-video — an installable skill for AI agents, published by heygen-com/skills.