Skip to content
SkillScript-led5,850 tokens in use

onetake

Makes 10 to 60 second product launch films and narrated feature demos in a single HTML composition, with measured motion, real motion blur, synthesized sound and a check that fails slideshows.

4,283 stars on the repository Custom / Unknownfeitangyuan/onetake
git clone https://github.com/feitangyuan/onetake.git ~/.claude/skills/onetake
View source

Install command as written in the repository's README

Skill label

measured from its files

What installing onetake costs

373tokens idle
5,850tokens in use
66,767tokens, up to

Idle is what sits in context while the skill is installed. In use is its instructions, loaded when it runs. Up to adds the files its instructions link. Counts are approximate, made offline with cl100k_base.

Model dependence Script-led

Script-ledMixedModel-led

Most of this skill is scripts, so it behaves much the same whichever model runs it.

This describes the skill's structure, how much of it is scripts. It does not say how well it works. Scripts: lib/motion.js, lib/ui_kit.js, scripts/analyze_ref.py, scripts/frame_diff.py, scripts/gallery.py, scripts/look.py, scripts/probe.py, scripts/record_footage.py, scripts/render.py, scripts/sfx_palette.py, scripts/stills.py, scripts/test_motion.js, scripts/verify_promo.py, scripts/vo_tools.py, templates/sfx_score.example.py.

Touches

shelldetectednetworkdetectedwrites filesdetected

Declared tags come from the skill's frontmatter; detected ones from its scripts, its code blocks and the commands it names in inline code. This is detected, not audited.

How it scores

Quality

How well the skill is written, counted from its own files: its frontmatter, its steps, examples, stated output and constraints, its size and whether every file it ships is one it names.How we score it →
Good
  • Frontmatter7/15
  • Structure11/25
  • Examples15/15
  • Output15/15
  • Size9/15
  • Resources14/15

Security

Starts at 100 and loses points for each thing in the skill's files that a reader should look at first.How we score it →
Good
  • Runs shell commands−5

    does this on the machine, and the frontmatter does not say so

    SKILL.md, line 44
  • Writes files−8

    does this on the machine, and the frontmatter does not say so

    SKILL.md, line 115
  • Reaches the network−10

    does this on the machine, and the frontmatter does not say so

    scripts/look.py, line 105

Repository health

Pre-computed score based on development activity, maintenance, community, maturity, and trend momentum.How we score it →
Good
  • Development activity84
  • Maintenance84
  • Community48
  • Maturity12
  • Momentum40

onetake builds short motion films about a product from the product itself. The agent deconstructs a reference clip into numbers, proposes three concepts that differ in their central picture, writes a beat sheet for rhythm and carry, and composes one HTML file where every value is a function of time, using a library of measured moves.

Frames are rendered with shutter motion blur at 1080p30 for drafts and 4K60 for finals, sound is synthesized in one shared room, and a verification script fails films whose beats replace each other. The repository is licensed for noncommercial use, with a separate paid commercial option.

What You Get

  • A single HTML composition driven by a seek function, built on a library of about 39 measured moves
  • A draft render at 1080p30 and a final render at 4K60 with real motion blur
  • A synthesized or recorded sound score timed from the composition’s own events
  • Optional narration with word timings, subtitles and extra languages from the same composition
  • A verification report on rhythm, continuity, motion curves and framing

Common Use Cases

  • Making a 15 second launch video for an app, a page or an agent skill
  • Building a narrated feature demo from screenshots without recording the screen
  • Producing the same film in several languages from one composition
  • Finding out why an existing motion video feels like a slideshow and fixing it

What Makes It Unique

Most video tools stack scenes. This skill designs the boundaries and measures them: a probe script scores what carries across each cut and the film fails when beats simply replace each other. It also forbids defaulting to a template story or a previous look, requires three distinct concepts, and ships case studies recording what each rejected cut taught.

Under The Hood

The skill ships Python scripts, a JavaScript motion library and templates. The agent runs Python scripts that drive headless Chromium with Playwright to record real pages and capture frames, uses ffmpeg to encode video, and renders in parallel browser processes. Narration uses local speech tools, and recorded sound sources are downloaded separately by the user. It writes footage, stills, audio and video files to your working folders, and the package it needs must be installed on your machine.

Join founders building with open source

Opinionated takes, migration guides, cost-saving tips, and insights from the open source ecosystem.

Subscribe on SubstackJoin 750+ subscribers