@heygen-com
Covers multimedia, frontend, content creation; known for video-edit, hyperframes, hyperframes-cli.
All Skills by @heygen-com
hyperframes-cli
by @heygen-com
CLI tool for the HyperFrames framework, provides scaffolding, preview, build, and export for a complete workflow.
hyperframes
by @heygen-com
HTML-based video framework, uses `data-*` attributes to control timeline and animations, automatically handles clip synchronization and media playback.
media-use
by @heygen-com
This skill acts as an Agent Media OS, resolving any media need (BGM, SFX, image, icon) into a frozen local file and a ledger record. A single `resolve` command handles the full cascade: project cache, global cache, HeyGen catalog search, freezing, and registration. It efficiently manages media assets, reducing search noise on disk for the agent and providing a clear file path.
hyperframes-animation
by @heygen-com
This skill provides comprehensive animation knowledge for HyperFrames, encompassing atomic motion rules, multi-phase scene blueprints, transitions, and broader motion-design techniques. It integrates seven runtime adapters such as GSAP, Lottie, and Three.js, allowing developers to efficiently compose complex animations, manage scene orchestration, and implement diverse motion designs across various platforms, ensuring deterministic and seek-safe animation experiences.
hyperframes-registry
by @heygen-com
HyperFrames component registry, provides reusable video blocks and effect components, installed and used via commands.
hyperframes-core
by @heygen-com
This skill defines the technical contract and build guide for HyperFrames video composition projects. It details how to construct renderable video projects using HTML, `data-*` attributes, tracks, and sub-compositions, ensuring deterministic composition structure, timing synchronization, and media playback. Developers must read this guide before writing composition HTML to ensure projects comply with framework specifications and render efficiently.
hyperframes-creative
by @heygen-com
This skill provides non-animation creative direction for HyperFrames videos, focusing on design specifications, brand guidelines, visual palettes, typography, and composition patterns. It helps define the aesthetic and narrative structure, including narration, beat planning, and audio-reactive visuals, ensuring consistent style and high-quality visual output for video projects. Use this skill to establish the foundational creative elements before diving into animation.
hyperframes-keyframes
by @heygen-com
This skill defines a contract and best practices for authoring web keyframe animations, ensuring visible states, continuous subject identity, seek-safe runtime, and verified pixels. It helps frontend developers write maintainable animation code by providing rules for GSAP, CSS, Anime.js, and WAAPI, avoiding common pitfalls like non-deterministic timing or unregistered timelines.
hyperframes-audio
by @heygen-com
This skill handles voice-over and music bed balancing in audio mixes. It analyzes spectral collisions between speech and music, then generates dynamic ducking envelopes and EQ settings written directly as data attributes on the media element. The same Web Audio graph powers both live preview and offline render, so what you hear is what gets written. Ideal for podcasts, video narration, and short-form content, it cuts mix time dramatically while keeping the music bed present.
general-video
by @heygen-com
This skill provides a general video workflow and custom composition authoring capabilities, serving as a fallback for specialized video production. It guides users through the entire process from visual identity and design systems to specific composition, ensuring professional and brand-consistent video content. This Skill is suitable for scenarios requiring flexible creation of various video content, with an emphasis on visual details and brand guidelines, helping users efficiently produce high-quality videos.
product-launch-video
by @heygen-com
This skill transforms a product or marketing URL, a pasted script, or a brief into a comprehensive product launch video. It covers various formats such as SaaS promotions, feature reveals, app launches, company promos, and product marketing videos. Users can leverage this skill to efficiently generate high-quality marketing videos for promoting, launching, or revealing new products, significantly enhancing market outreach and engagement. It streamlines the video creation process, allowing for rapid production of compelling visual content.
faceless-explainer
by @heygen-com
This skill transforms any text, such as articles or notes, into a faceless explainer video, typically 30-90 seconds long. It invents all visuals—including typography, abstract graphics, diagrams, and data visualizations—without relying on external assets or website captures. Ideal for creating engaging topic explainers, concept breakdowns, how-tos, and narrative content, it provides a unique way to convey information visually when no real-world footage is available.
motion-graphics
by @heygen-com
This skill specializes in generating short, design-led, unnarrated motion graphics. It automates the process of transforming text, data, or brand elements into captivating visual animations, eliminating the need for professional design or video editing expertise. This enables efficient information delivery and enhances content appeal.
remotion-to-hyperframes
by @heygen-com
This skill translates existing Remotion (React-based) video compositions into HyperFrames (HTML) compositions. It handles most Remotion idioms, identifies patterns that cannot be losslessly translated, and recommends runtime interop solutions to ensure translation quality.
embedded-captions
by @heygen-com
This skill intelligently adds embedded captions to talking-head videos, supporting various visual styles and special effects to achieve cinematic or VFX-level captioning. It automates the process of deeply integrating captions with video content, significantly enhancing the video's visual appeal and professionalism, especially for scenarios requiring advanced dynamic captions.
hyperframes-studio
by @heygen-com
This Skill defines timeline-organization conventions for video projects: every scene becomes its own sub-composition, all captions live on a single track, each element kind gets its own track, and content respects action-safe and title-safe zones. The result is a short, readable timeline where creators can trim, mute and drag each clip as a unit instead of facing a wall of opaque rows, and a lint pass confirms the structure before delivery.
pr-to-video
by @heygen-com
This skill automatically transforms a GitHub Pull Request into a code-change explainer video, up to 3 minutes long. By analyzing diffs and commit history, it generates clear changelog, feature reveal, or refactor walkthrough videos, helping developers and teams efficiently communicate code changes, enhancing understanding and collaboration.
music-to-video
by @heygen-com
This skill transforms a music track into a beat-synced HyperFrames video. It analyzes the music's beats and energy to intelligently orchestrate video frames, integrating optional user-supplied images or videos. This automates the creation of high-quality, rhythmically engaging visual content, significantly streamlining the video production process.
talking-head-recut
by @heygen-com
This skill packages an existing talking-head, interview, or podcast video with timed, designed graphic overlays (titles, lower-thirds, data callouts, quotes, side panels, picture-in-picture) synced to the transcript. It eliminates the need for professional video editing software, enabling content creators, marketers, and podcasters to produce polished, broadcast-ready videos quickly.
slideshow
by @heygen-com
This skill generates structured, navigable HyperFrames slideshows, incorporating scene definitions, speaker notes, fragment control, and branching logic. It addresses the pain points of time-consuming and less interactive traditional presentation creation. Users can efficiently produce professional and expressive interactive presentations, suitable for product pitches, technical sharing, and various other scenarios.
hyperframes-media
by @heygen-com
Asset preprocessing for HyperFrames compositions — text-to-speech narration (Kokoro), audio/video transcription (Whisper), and background removal for transparent overlays (u2net). Use when generating voiceover from text, transcribing speech for captions, removing the background from a video or image to use as a transparent overlay, choosing a TTS voice or whisper model, or chaining these (TTS → transcribe → captions). Each command downloads its own model on first run.
website-to-video
by @heygen-com
This skill captures any general website or URL and transforms it into a professional HyperFrames video, ideal for creating site tours, showcases, or social media clips. It leverages headless Chrome screenshots and brand assets to generate engaging video content directly from the website's own visuals. This skill significantly streamlines the process of converting static web pages into dynamic video, helping users quickly produce high-quality visual marketing materials to enhance content appeal and dissemination efficiency.
gsap
by @heygen-com
A comprehensive GSAP animation reference guide tailored for the HyperFrames platform. It details core animation methods like `gsap.to()`, `from()`, `fromTo()`, along with easing, staggering, timeline control, and performance optimization techniques. Developers can leverage this guide to efficiently create smooth, professional animations, significantly enhancing the visual appeal and interactivity of their HyperFrames compositions.
website-to-hyperframes
by @heygen-com
Automatically captures websites and generates professional product videos, supports custom durations and platform formats. An AI Agent Skill, enhancing work efficiency and automation capabilities.
css-animations
by @heygen-com
CSS animation adapter patterns for HyperFrames, enabling deterministic seeking of CSS keyframes, animation-delay, and fill-mode for preview and rendering. Ideal for simple, fixed-duration decorative motions like loops, masks, and subtle parallax.
tailwind
by @heygen-com
This skill provides essential patterns for integrating Tailwind CSS v4.2 browser-runtime into HyperFrames compositions. It guides users on scaffolding projects with `hyperframes init --tailwind`, writing utility classes, defining CSS-first v4 theme tokens, and navigating v3 vs v4 syntax differences. It also helps in debugging missing styles and understanding when to compile Tailwind to CSS for production-stable renders, ensuring deterministic and reliable styling.
animejs
by @heygen-com
This skill provides Anime.js adapter patterns for HyperFrames, enabling precise control over animations and timelines. It ensures seek-driven, deterministic playback and supports translating Anime.js examples into render-safe HyperFrames HTML, simplifying animation integration.
three
by @heygen-com
Three.js and WebGL adapter patterns for HyperFrames. Use when creating deterministic Three.js scenes, WebGL canvas layers, AnimationMixer timelines, camera motion, shader-driven visuals, or canvas renders that respond to HyperFrames hf-seek events.
lottie
by @heygen-com
Lottie and dotLottie adapter patterns for HyperFrames. This skill facilitates embedding lottie-web JSON animations and .lottie files, ensuring deterministic playback and precise control. It enables seamless synchronization of animations for logo reveals, UI motion, and After Effects exports within HyperFrames, significantly enhancing frontend animation development efficiency and quality.
waapi
by @heygen-com
Web Animations API adapter patterns for HyperFrames. Use when authoring element.animate() motion, Animation currentTime seeking, document.getAnimations(), KeyframeEffect timing, fill modes, or native browser animations that must render deterministically in HyperFrames.
motion-doctrine
by @heygen-com
This Skill is the motion-law gateway for HyperFrames, defining a unified vector law (axis, direction, speed, phase) for multi-scene videos to ensure seamless transitions that feel like one continuous camera move. It eliminates visual breaks and idle wobble, ideal for animation automation, video transition consistency, and cinematic flow.
captions-overlay
by @heygen-com
This Skill provides a caption overlay doctrine for video production, defining three caption modes (drop/rail/embed) and the rule that captions are an overlay composited on top of the film, not a reserved band. It prevents common mistakes like over-embedding and shifting content away from center, ensuring professional captioned compositions for talking-head, launch, and explainer videos.
cut-the-curve
by @heygen-com
This skill provides a complete catalog of motion graphics transition techniques, including five velocity-matched SEAM transitions (zoom-through, inverse zoom-through, cut-the-curve, waterfall cut, rack-focus blur-cut) and two in-scene techniques (waterfall entry, nudge curve). It enables video editors and motion designers to create cinematic transitions with matched velocity and direction across cuts, enhancing visual smoothness and professional quality.
seam-craft
by @heygen-com
This Skill enforces render-correctness for scene-to-scene transitions in video compositing, preventing white flashes caused by transparent backgrounds during opacity dips. It automates the application of stage background, overlap mechanics, and template stamping on the master timeline, ensuring smooth and professional transitions without manual per-scene adjustments. Ideal for video editors, render pipelines, and timeline assembly workflows.
changelog-video
by @heygen-com
This skill converts weekly changelog/digest Markdown into a finished branded changelog video (square 1080p, ~45-60s, with animated backgrounds, mock-UI visualizations, and lowkey captions). It is ideal for teams that need to produce regular product update videos without manual motion design or editing.
oversized-cursor
by @heygen-com
Generates oversized cursor animations that guide viewer's eye, trigger scene changes, and add motion to launch videos. Uses physically believable off-screen entry, click ignition, and exit to provide cheap high-yield visual dynamics.
contribute-catalog
by @heygen-com
This skill guides users on how to contribute new registry blocks (e.g., caption styles, VFX blocks) or components (e.g., text effects, overlays) to the HyperFrames repository. It covers the complete workflow from idea generation to PR merge, ensuring contributions comply with specifications.
typegpu
by @heygen-com
TypeGPU and raw WebGPU adapter patterns for HyperFrames. Use when creating GPU-rendered compositions with TypeGPU, raw WebGPU, WGSL fragment shaders, compute pipelines, liquid glass effects, particle systems, or any canvas layer driven by navigator.gpu that responds to HyperFrames hf-seek events.
graphic-overlays
by @heygen-com
This skill enhances existing talking-head or interview videos by layering timed, designed graphic overlay cards such as titles, lower-thirds, and data callouts. It automatically designs and generates HTML for each card based on the video transcript, then renders the final composition to MP4. This significantly streamlines the post-production process of adding professional visual elements, boosting video engagement and polish.
heygen
by @heygen-com
This skill is deprecated and was originally used to generate video content based on prompts. Please use the `create-video` skill instead for video generation functionality.
video-translate
by @heygen-com
Specializes in video content translation and multi-language dubbing, capable of rapidly localizing existing videos into multiple languages to expand global audience reach.
liveavatar-integrate
by @heygen-com
Scans an existing codebase to detect the user's stack, then recommends and implements the single best LiveAvatar integration pathway — Embed, FULL, or LITE mode, including custom LLM, custom TTS, ElevenLabs agent, and bring-your-own WebRTC variants. Instead of reading pages of docs and trying paths that don't fit, developers get one clear recommendation and a guided build so a lip-synced, eye-contacting real-time video avatar ships in their product quickly.
video-understand
by @heygen-com
Understand local video content by extracting frames with ffmpeg and transcribing with Whisper, running completely offline without API keys.
avatar-video
by @heygen-com
Create AI avatar videos with full control over avatar, voice, script, scene, and background, using HeyGen /v2/video/generate API to build single or multi-scene videos.
ai-video-gen
by @heygen-com
This skill empowers AI agents to generate realistic avatar videos using the HeyGen platform, complete with custom faces and voices. It transforms text instructions into engaging video content, perfect for creating personal introductions, messages, or dynamic agent communications. By automating avatar creation, style selection, and video generation, this skill dramatically streamlines the video production workflow, allowing agents to bring themselves to life and communicate more expressively.
create-video
by @heygen-com
This skill leverages the HeyGen platform to quickly create personalized AI avatar videos through an AI agent. Users only need to provide a photo and text to generate professional videos with voice and expressions, significantly boosting content creation efficiency.
video-download
by @heygen-com
Use yt-dlp to directly download videos and audio from URLs, supporting stream merging and audio track extraction, without additional script wrapping.
visual-style
by @heygen-com
This skill empowers AI agents to create personalized avatar videos. It can generate an avatar from a photo and then use it to produce short introductory videos or other video messages. It streamlines the process of bringing digital identities to life, providing a face and voice for AI agents to communicate visually, thereby enhancing agent interaction and content delivery.
faceswap
by @heygen-com
Use HeyGen API for AI face swapping, replacing faces from source images into target videos, supporting GPU-accelerated processing.