Hyperframes To After Effects logo

Hyperframes To After Effects

Organization
Clipwing
hyperframes-to-after-effects

Convert a HyperFrames project (HTML/GSAP video compositions) into an editable Adobe After Effects project (.aep) with real layers and keyframes, not a rendered video. Text stays live text, cards become shape layers, images and video become footage, sub-compositions become precomps, animation becomes keyframes, and the result is checked against the HyperFrames render frame by frame. Use this whenever someone wants to move, port, export or "hand off" a HyperFrames / HTML video to After Effects or AE, wants an .aep from their HyperFrames composition, wants to keep editing a HyperFrames film in After Effects, or asks why an AE import of their HTML animation is slow or broken, even if they do not say "skill" or name the tool. Windows and macOS.

Overview

PublisherClipwing
Repositoryskills
Skill namehyperframes-to-after-effects
Stars
55
Forks
1
Bundled files
12
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 12 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by Clipwing on GitHub. Read the source before you install it.

Installation

Install the Hyperframes To After Effects AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/Clipwing/skills/tree/main/skills/hyperframes-to-after-effects
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/Clipwing/skills.git /tmp/skills
mkdir -p .claude/skills
cp -r /tmp/skills/skills/hyperframes-to-after-effects .claude/skills/hyperframes-to-after-effects
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Hyperframes To After Effects in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Hyperframes To After Effects on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Hyperframes To After Effects is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

HyperFrames to After Effects

A HyperFrames film is a paused GSAP timeline over HTML. This skill plays it in headless Chrome frame by frame, records what every element looks like, and rebuilds that in After Effects as native, editable layers:

HyperFramesAfter Effects
sub-composition (data-composition-src, nested data-composition-id)precomp at its start, trimmed to its visible window
text node (each rendered line)text layer, live font; changing strings become native recipes (below)
box with background / border / radius / outline / hard shadowshape layer (rounded rect, fill, stroke, dashes)
soft box-shadow / text-shadow, filter: blur(), mix-blend-modeDrop Shadow, Gaussian Blur (keyed), blending mode
gradients, background images, inset or multiple shadows, mixed radiiPNG painted by Chrome (transform and opacity still keyed)
<img> (object-fit, rounded corners), <video> (trim, rate, volume)footage layer with fit, crop mask, timing, audio level
inline <svg>PNG at 4x (a few looks: one layer each; many: a sequence)
<canvas> (2D, three.js, Lottie)PNG sequence with time remap
overflow: hidden, clip-path (polygon, inset, circle, ellipse)track matte (nested clips nest)
<audio>audio layer at its start, with its volume

Changing text never becomes per-frame Source Text keys: After Effects slows down project-wide once any layer has them (one 150-key layer makes every later scripted edit ~200x slower, a 3,000-key film took hours). Instead: counters → a Count slider + expression (or pure time expression when the number follows time), typing → Text Animator, a few strings or captions → one trimmed layer per string, scrambles → a small [time, text] expression table, colour changes → Fill Color animator.

Before you start

Run the doctor and read it to the user in plain words:

bash
node <skill>/scripts/hf2ae.mjs doctor <project>
  • Node 18+, the project's HyperFrames CLI and its Chrome. The tool reuses what HyperFrames already installed (puppeteer-core, fontkit, sharp, Chrome) from the npx cache. If Chrome is missing, npx hyperframes browser ensure downloads it: ask the user before running it.
  • After Effects (any recent version; 2023+ for the track-matte API) with Preferences → Scripting & Expressions → Allow Scripts to Write Files and Access Network switched on. Ask the user to check it; the build cannot set it.
  • ffmpeg for verification (HyperFrames rendering needs it anyway) and for converting WebM/OGG media.

Ask the user (only what you cannot infer):

  1. Where to save the .aep (default <project>/after-effects/project.aep).
  2. If After Effects is open with a project: build into that open project (--into-open), or save and close it first. Never close, save over or discard the user's open project yourself.
  3. Frame rate: the project's render rate if known (hyperframes render -f), else 30. Use 60 for fast UI motion.

Workflow

bash
node <skill>/scripts/hf2ae.mjs export <project> --out <dir> [--fps 30]   # Chrome only, no After Effects
node <skill>/scripts/hf2ae.mjs build  --out <dir> [--aep <file>] [--into-open]
node <skill>/scripts/hf2ae.mjs verify <project> --out <dir> [--frames 8]
node <skill>/scripts/hf2ae.mjs all    <project> --out <dir> ...           # the three in a row

<skill> is this skill's folder. Run steps one at a time the first time on a project, so problems surface at the step that caused them.

  1. Export writes <dir>/manifest.json, <dir>/comps/*.json, <dir>/assets/, <dir>/fonts/ and <dir>/export_report.txt. Read the report before building: its fonts list says which real face every CSS family resolves to (aliases such as "Brand Sans" are read from the font files), and warnings lists what was approximated. Tell the user about warnings that affect what they will see.
  2. Build starts After Effects (or uses the open one), writes progress to <dir>/ae_status.txt and a log to <dir>/ae_build_log.txt, saves after every composition. Typical films build in under a minute. If it runs for minutes on one layer, stop and read the log rather than waiting: see references/troubleshooting.md.
  3. Verify renders the .aep with aerender and puts After Effects and HyperFrames side by side in <dir>/verify/compare.png, with SSIM per moment in <dir>/verify/report.txt. Open the sheet and look at it. SSIM tells you where to look, your eyes tell you what is wrong: typical numbers are 0.98+ with the original fonts and 0.95–0.99 with substitutes. Fix causes in the export or build (not by hand in After Effects), rebuild, re-verify.

Fonts

After Effects only sees fonts installed for all users (on Windows a per-user install is invisible to it). The build uses the real face when After Effects has it; otherwise it picks the closest installed face of the same kind (sans / serif / mono), fits each line to the width it had in the film (so words placed side by side keep their spaces), and writes the original font into the layer comment. The files to install are in <dir>/fonts/ (Google Fonts are fetched as TTF). Offer the user that choice: install them for all users, restart After Effects, run build again (no re-export needed). Installing fonts is the user's action, not yours.

Reporting back

Tell the user, briefly: where the .aep is; what became what (precomps, text, shapes, footage); which fonts were substituted and how to get the originals; the warnings that matter; the verify result with the sheet. Mention what stays approximate (3D transforms flattened to 2D, CSS filters other than blur, painted PNGs for gradients, blend modes that After Effects mixes slightly differently) so nothing surprises them in After Effects.

Going deeper

  • references/how-it-works.md: the export format, keyframe reduction, text recipes, mattes, timing of precomps and media. Read it before changing the scripts.
  • references/troubleshooting.md: slow or stuck builds, After Effects not running the script, fonts, mismatched frames, macOS specifics.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Hyperframes To After Effects AI skill do?

Convert a HyperFrames project (HTML/GSAP video compositions) into an editable Adobe After Effects project (.aep) with real layers and keyframes, not a rendered video. Text stays live text, cards become shape layers, images and video become footage, sub-compositions become precomps, animation becomes keyframes, and the result is checked against the HyperFrames render frame by frame. Use this whenever someone wants to move, port, export or "hand off" a HyperFrames / HTML video to After Effects or AE, wants an .aep from their HyperFrames composition, wants to keep editing a HyperFrames film in...

Why use Hyperframes To After Effects on TypingMind?

Because you install it once and use it with any model. Hyperframes To After Effects is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Hyperframes To After Effects in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/Clipwing/skills/tree/main/skills/hyperframes-to-after-effects. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Hyperframes To After Effects?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Hyperframes To After Effects?

As many as you like. As long as a model supports skills, you can use Hyperframes To After Effects with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Hyperframes To After Effects AI skill free?

It is published on GitHub by Clipwing. Check the repository for licensing terms. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇