website-to-video

Capture a website and produce a HyperFrames video from its brand assets.

Updated Jul 16, 2026
One-click install
npx skills add https://github.com/X-manist/Cohmira --skill website-to-video-x-manist
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: website-to-video
Source: https://github.com/X-manist/Cohmira/tree/main/src/builtin-plugins/openmontage/.agents/skills/website-to-video
Command: npx skills add https://github.com/X-manist/Cohmira --skill website-to-video-x-manist

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Turning a website into a polished video normally requires manual screenshotting, brand asset extraction, storyboarding, voiceover recording, and frame-by-frame animation work. This Skill automates that pipeline by capturing a URL with headless Chrome, extracting the site's brand identity, and guiding a gated 7-step workflow that produces a HyperFrames video composition. ## Core Features & Use Cases - Automated Site Capture: Runs npx hyperframes capture to download screenshots, SVGs, fonts, design tokens, and computed CSS styles from any URL. - Brand Identity Extraction: Generates a DESIGN.md cheat sheet with exact colors, font files, component CSS values, and WCAG contrast checks. - Storyboard, Voiceover, and Build Pipeline: Creates storyboards and narration scripts, generates TTS audio with word-level timing, and builds GSAP-based beat compositions validated by lint and snapshot gates. - Use Case: A marketer asks to turn the company homepage into a 15-second Instagram clip; the Skill captures the site, locks the message and narrative arc, builds branded animated beats, and delivers a Studio preview URL. ## Quick Start Ask the agent to turn a specific website URL into a short social video, for example: make a 15-second Instagram clip from https://example.com.

Frequently Asked Questions about website-to-video

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I turn a website into a video?

Run the capture step with npx hyperframes capture on the target URL, then follow the 7-step workflow: extract brand identity into DESIGN.md, lock the message and storyboard, generate narration, build GSAP beat compositions, and validate with lint and snapshots before delivery.

What video formats and durations does website-to-video support?

It supports landscape 1920x1080, portrait 1080x1920, and square 1080x1080 outputs. Typical durations range from 10-15 second social ads to 30-60 second product demos, with beat count driven by the storyboard rather than fixed targets.

When should I use product-launch-video instead of website-to-video?

Use product-launch-video whenever the intent is marketing, launching, or promoting a product, even if the source is a URL. Website-to-video is only for general site tours, showcases, or social clips made from the site's own visuals.

Does the capture step require an API key?

The base capture works without any API key. Setting a Gemini API key is optional and adds AI-generated descriptions of captured images, which improves asset selection for each scene at roughly $0.001 per image.

Why does my video composition fail hyperframes lint?

Common causes include missing window.__timelines registration, mismatched data-composition-id attributes, non-deterministic code like Math.random or requestAnimationFrame, and gsap.from with CSS opacity zero. Fix all lint errors before snapshotting and delivery.