auto-compare-video skill
The create-video skill builds short vertical "X vs Y" videos on HyperFrames with TTS voiceover for TikTok, Reels and Shorts.
- 228 ★ on the repository
- Last commit
Extended access
Why this level
- The skill runs npm scripts and FFmpeg, and voice services receive texts through keys from `.env`.
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/auto-compare-videoDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- The skill runs npm scripts and FFmpeg, and voice services receive texts through keys from `.env`.
This is third-party code. Review the repository files before installing.
What it does
The repository defines a series of educational videos where each one compares two easily confused ideas. The layout is fixed: two concept cards on top, line-by-line captions in the middle and a robot host below. Only the content changes, the design is described in `DESIGN.md`. On a request like "make a comparison video of X and Y" the `create-video` skill creates a `videos/<slug>/` folder with a ready HyperFrames project: a 12-line script for 30–40 seconds, Vietnamese voiceover and animation timed to the measured audio length. Then `npm run check` and `npm run render` produce an MP4. The voice comes from VieNeu TTS (local, free, needs CIT Voice Studio at `http://127.0.0.1:8001`), Edge TTS (free, no key) or Vbee TTS (account required). Two finished videos in the repository serve as structure samples.
Who it is for. Creators of short Vietnamese educational videos who want to release a series in one layout through Claude Code, Cursor or Codex.
Good fit when
- A series of "term A vs term B" videos goes out to TikTok, Reels and Shorts.
- The voiceover has to be free: Edge TTS needs no key.
- You want a HyperFrames project sample with voiceover and audio length measurement.
- A new video has to join a ready series without reworking the layout.
Not a fit when
- The video has another format: a product promo, a video from a URL, a slideshow or subtitles for existing footage.
- The voiceover must be in a language other than Vietnamese: voices and texts are built for it.
- The series layout is meant to change from video to video.
Example request
Make a comparison video "Dev vs DevOps" in the series format with Edge TTS voiceover.Limitations
Node.js 18 or newer, FFmpeg and FFprobe in `PATH` and one of the voice services are required. Both finished videos use the old 8-line, 15–20 second format while the skill requires the new 12-line format, so only the structure can be copied from them. Publishing to channels needs keys in `.env`, which must not be committed. GitHub does not detect the license file automatically, the README states MIT. The last commit dates from September 2026.
How to disable. Delete `.claude/skills/create-video` (and the copy in `.agents/skills/`) from the clone and restart the agent. Finished videos sit in `videos/` and are deleted separately.
FAQ
Which voice should I use without keys?
Edge TTS: it is free, needs no key and is already wired in through the `edge-tts-universal` npm package.
How long is a video?
30–40 seconds with a 12-line script. Timing is fitted to the measured voiceover length.
How do I render a video?
In `videos/<slug>/` run `npm run check`, then `npm run render` to get an MP4.
Which agents fit?
The skill is described for Claude Code, Cursor, Codex and other agents that read the SKILL.md format.
What must be installed beforehand?
Node.js 18 or newer, FFmpeg and FFprobe in `PATH`, and the chosen voice service for narration: VieNeu, Edge TTS or Vbee.
Discussion
No questions or reviews yet.
Related
SkillsThe video-explainer skill turns a script into a vertical 9:16 explainer: narration, captions, Remotion and a pre-render review
A skill that turns a messy phone recording into a vertical reel: best takes, snap zooms on cuts and word-timed captions.
A Claude Code skill for narrated explainers with a cartoon presenter, lip sync to your voice, poses on cue words and your brand
A skill that turns a topic into a narrated 4K video: research, script, local TTS, Remotion composition and an FFmpeg mix
A skill that turns your narration and real assets into a 20–60 second video in the style you pick, rendered locally as MP4.
A skill that turns an idea, a link or a screen recording into finished video: HeyGen avatar, b-roll, captions and quality checks
Comparison with similar tools
Cards that share the main search query, and how they differ.
| Tool | Official | Repository stars | Main difference |
|---|---|---|---|
| auto-compare-video (this card) | no | 228 | |
| HyperFrames | no | 83,841 | HyperFrames is the composition engine, while auto-compare-video adds a ready series, layout and voiceover pipeline. |
| HyperFrames Student Kit | no | 2,795 | Student Kit teaches HyperFrames basics, while this repository releases finished videos in one layout. |
README in short
A template for a series of short comparison videos on HyperFrames with a fixed layout, Vietnamese voiceover and the `create-video` skill for new episodes.