general-videoby HeyGen

General Video Skill.

A Claude Code and Codex skill for motion and video: authors or edits a custom multi-scene HyperFrames video when no template fits.

Motion & Video
Video

Download OpenDesign View usage guide

heygen-com/hyperframes · skills/general-video/SKILL.mdSKILL.md

--- name: general-video description: 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, brand and sizzle reels, montages, static loops, static title cards, footage remixes, and freeform builds. Use motion-graphics instead for a short unnarrated motion-first unit, including an animated title. Route fresh creation through hyperframes before using … --- ## 1. Apply cross-cutting source adapters ## 2. Start from project state ## 3. Interpret the run shape ## 4. Load required knowledge before each stage ## 5. Execute the composition ## 6. Gates that always apply

Front matter and section headings read from the upstream file on 2026-09-14 · 20,760 characters in full

01 What it does

What the general-video Skill Does in Claude Code and Codex

Summary, workflow and outputs, all read from the skill's SKILL.md.

From the SKILL.mdRead 2026-09-14 · not a recorded run

In the Skill's Own Terms

Authors or edits custom HyperFrames compositions for multi-scene videos, brand reels, montages, footage remixes, static loops, title cards, and freeform builds when no specialized workflow template fits. Supports both automation flow (agent chooses and executes the route) and companion flow (co-create in conversation with director-level treatment planning).

Handles scene planning through storyboarding, multi-scene assembly with optional worker dispatch for scale, media operations integration, and maintains deterministic composition contracts. Used for longer pieces or custom builds where shipped workflows don't apply.

What it produces

compositions/<frame_id>.html for each scene with timed clip elements, sized root, and one paused seek-safe timeline on window.__timelines
compositions/<frame_id>.motion.json sidecars carrying durations and exit/entry vectors for each scene
STORYBOARD.md with one Frame N block per scene containing status, src, blueprint/rules citation, and beat text
Rendered video file verified after user approval of final Studio preview

See the example prompt ↗

How It Works

01
Apply Source Adapters
Load media-use for any audio, image, voice, grade, or caption needs. If input is a Figma URL, run the Figma skill first to export assets, tokens, and storyboard frames.
02
Start from Project State
For specific edits, make the change and preserve decisions without reopening discovery. For fresh creation, run hyperframes intent layer first.
03
Interpret Run Shape
Read flow (automation or companion), storyboard (yes or no), and derived mode from BRIEF.md. Automation chooses the route; companion co-creates.
04
Execute Composition Stages
Plan the viewer arc with cited blueprints or rules. Search the live catalog for any named look before building it. Resolve dependencies and install registry blocks.
05
Assemble and Verify
Mount scenes, media, transitions, captions, and audio. Carve music beds under voice tracks. Run lint after first HTML pass, then hyperframes check for the final gate.
06
Approve and Render
Open final Studio preview only after checks pass. Ask whether to render or revise. Render only after user approval.

02 Find your fit

When to Use the general-video Skill

Use the general-video skill in Claude Code or Codex when the job is motion and video of this kind. Fit and limits below are taken from the skill's own file.

Good Fit

Longer or multi-scene pieces including brand and sizzle reels, montages, static loops, static title cards, and footage remixes
Custom freeform builds when no specialized workflow template fits the project requirements

Know the Boundaries

Short unnarrated motion-first units including animated titles should use motion-graphics workflow instead
Fresh creation requests should route through hyperframes intent layer first, not directly to this skill

What to Provide

BRIEF.mdWritten using brief-format.md when it exists. Contains workflow, flow, storyboard preferences, and confirmed fields.
Figma URLsAny figma.com URL input triggers the Figma skill first to export assets, tokens, components, or storyboard frames.
Existing project stateResume from hyperframes.json or STORYBOARD.md when they exist. Backfill BRIEF.md only from known facts without asking brief questions.

03 Inside the skill

Rules the general-video Skill Gives Claude Code and Codex

5 concrete instructions, defaults and limits the SKILL.md sets for the agent — the part of the file that changes the result.

Search Catalog Before Building Looks
Before planning to build any named look yourself, run npx hyperframes catalog --query with the look in plain English.
Inline Faster Than Dispatch at Small Scale
Up to approximately 6 short scenes build faster inline in one context (measured: 5 scenes approximately 9 minutes inline vs approximately 21 minutes packetized).
Hard Cuts Use Duplicate Sources
For hard cuts, trims, splices, or reorders of existing footage, duplicate the same video source into multiple clip elements.
Build Exact Scope Only
Build exactly what the user asked for. A title card is not a title card plus three scenes, music, and captions. Offer additions before adding them.
Design Before HTML
Establish design source in order: frame.md then design.md then DESIGN.md.

04 Put it to work

Install general-video in Claude Code or Codex

One npx skills add command, then a first task in your agent and a checklist for the result.

Add general-video to Claude Code, Codex or Your Agent

Run in your project; the installer asks which agent to add it to.

Terminal

npx skills add https://github.com/heygen-com/hyperframes --skill general-video

Generic skills CLI form for skills/general-video/SKILL.md; the repository may document its own path.

Give the Skill a First Motion and Video Task

Written in the skill's own terms; replace the bracketed parts with your material.

Starter prompt

Create a 90-second brand sizzle reel for [company name] using [uploaded footage folder]. Open with their logo reveal, follow with four product highlight scenes showing [feature A], [feature B], [feature C], and [customer testimonial clip], then close with a call-to-action card displaying their tagline and website. Use their brand colors from [design.md] and add background music that feels energetic and modern. Include lower-third captions for the testimonial.

Source: written from the SKILL.md in the skill's own terms.

Check the First Result

What the SKILL.md itself says a correct result looks like:

Run npx hyperframes check for the final gate, which reruns lint internally plus additional validations
Inspect midpoint snapshots for sub-compositions to verify scene rendering
Review animation map using hyperframes-animation/scripts/animation-map.mjs for multi-scene work
Review design adherence against design-adherence.md when a design specification file exists

In OpenDesign import general-video from its SKILL.md link: Plugins → Add → Skill → Import from link, then pick it in the top bar.

Download OpenDesign Read the quickstart

05 Open source, traceable

Source, License and Verification

Where this page's facts come from, and when they were last checked.

Author: HeyGen
Repository: github.com/heygen-com/hyperframes Branch main
Skill file: skills/general-video/SKILL.md Content checksum recorded at verification; commit 558c11b
License: Apache-2.0 From LICENSE.
Repository stars: 49,761 GitHub snapshot taken 2026-09-14 for the whole heygen-com/hyperframes repository, which contains 9 skills in this catalog. Not a rating or usage count for general-video.
Verified by OpenDesign: 2026-09-14 Source file, path, license text and star count were checked. Not bundled in OpenDesign main (checked 2026-08-28); a live run has not been recorded on this page.

Inside the SKILL.md

General videoIntroduction and skill update command requirement before starting any work
1. Apply cross-cutting source adaptersLoad media-use for assets and operations; run Figma skill for any Figma URL inputs
2. Start from project stateMatch current state: specific edit, existing brief, resume from files, or fresh creation scaffold
3. Interpret the run shapeRead flow automation or companion, storyboard yes or no, and derived mode from brief contract
Companion flowArrive as director with ceiling treatment, upgrade channels between checkpoints, record accepted capabilities immediately
4. Load required knowledge before each stageMandatory reads table matching conditions to required references including core, creative, animation, storyboard, media, production
5. Execute the compositionSeven-stage dependency order: plan, review, resolve, build scenes, merge motion, assemble, verify, final approval
Keep scope exactBuild only what user asked for; offer additions before adding them
Establish design before HTMLResolve design source order, complete four items before writing HTML when no spec exists
Preserve the composition contractTimed elements use class clip, root sized, one paused timeline, deterministic rendering without network fetches
Borrow workflows safelyUpdate and borrow genre references and taste from similar workflows, not their private scripts or contracts
7. DoneRun complete checklist: scope implemented, checks pass, adherence reviewed, snapshots inspected, preview approved, render verified

The Skill's Own Description

“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, brand and sizzle reels, montages, static loops, static title cards, footage remixes, and freeform builds. Use motion-graphics instead for a short unnarrated motion-first unit, including an animated title. Route fresh creation through hyperframes before using this skill.”

Front-matter description from SKILL.md. The full file is about 20,760 characters. Read the full file on GitHub.

06 Before you install

Questions About the general-video Skill

Answers come from the SKILL.md as read on 2026-09-14, not from a recorded run.

Do I need a HeyGen account or sign-in to use this skill?

Authentication depends on the flow setting and required capabilities. Before the first authenticated provider action, the skill runs npx hyperframes auth status. For collaborative flow, it waits for sign-in or an explicit offline choice.

For autonomous flow, it states the status and continues through an available offline provider. Surface a blocker only when no offline provider can satisfy a required capability.

General Video

Run general-video on a Real Motion and Video Task.

Download OpenDesign, import general-video from its SKILL.md link, paste the prompt above, and read the plan before the result.

Download OpenDesign Browse all skills