agentleFS
vargHQvargHQ/skills/app-ui-mockups/SKILL.md

app-ui-mockups

Skill23 starsChanged 8 months agoMITView on GitHub

Ten animated UI mockups for product, launch and dev-tool videos, rebuilt as HTML instead of screen recordings: phone message thread, team chat channel, coding-agent terminal session, terminal command run, desktop terminal window, desktop notification banner, code editor walkthrough, 3D code slab, exploded UI panels and an amber CRT console. Presets give typing speed (12-20 characters per second with seeded gaps), reading holds from character count, entrance easing and scale, camera lead before each action, text size as a share of frame height, and theme colors from MIT theme sources, from Charm VHS, Screen Studio, Emil Kowalski, Apple HIG and Netflix timing rules. Use when someone wants a fake text or Slack conversation, a terminal or Claude Code demo, a code walkthrough or an exploded UI shot in a video. NOT for recording a real screen or for real people's messages presented as genuine.

The file

We split this skill into the rules it gives an agent

6 rules. About 0 are common, 4 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Invented people, real product: Use invented names and workspaces in chats.

    line 34
  2. Distinctive:

    Free fonts only: Every font in the presets is OFL or Apache.

    line 36
  3. Very specific:

    Seek-safe: Every value is a function of timeline time.

    line 37
  4. Distinctive:

    After a command hold 3 s (short output) / 5 s (long).

    line 71
  5. Distinctive:

    No platform logos; invented people and workspaces.

    line 77
  6. Very specific:

    Every rule in the references cites its source by number.

    line 83

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

App UI mockups: chats, terminals, code and notifications for launch videos

Ten UI scenes rebuilt as HTML so they look clean, read at a human pace and render the same every time. Typing, streaming and holds follow measured reading and typing speeds rather than guesses.

Gates

  • Invented people, real product. Use invented names and workspaces in chats; never put words in a real person's mouth or show a conversation as genuine when it is not.
  • No platform logos. UIs are recognisable look-alikes; the platforms' names, logos and UI looks are their owners' trademarks.
  • Free fonts only. Every font in the presets is OFL or Apache. Swap in the person's brand font only if they have a license for video.
  • Seek-safe. Every value is a function of timeline time; randomness is seeded by frame. No Date.now(), requestAnimationFrame or unseeded Math.random().
  • No credits here. These templates render locally for free. Generated footage, voice or music is the varg-ai skill: call estimate_cost and wait for the person's approval before any generation.

Pick a template

10 templates, 40 presets. Each reference file has the same sections: how the pros do it (every rule cited), presets, timing and numbers, how to build it, mistakes that look amateur, sources. Each JSON file holds the same presets as machine-readable values.

TemplateUse it forPresetsFiles
Coding-agent terminal sessionOne AI coding-agent turn (Claude Code-style) in a terminal: prompt, tool calls, diff, spinner, summary.Prompt and tool calls, Code edit, Still working, Huge typeagent-terminal-session.md · json
Amber booth consoleA retro amber projection-booth touchscreen boots, loads your show, counts down and draws your end card.Boot, load, end card, Clean amber, Footage to amber CRTamber-booth-console.md · json
Code editor walkthroughA VS Code-style guided tour: scroll to a line, highlight it, open a comment thread per step.Dark 2026, Light 2026, Editor only (vertical)code-editor-walkthrough.md · json
3D code slabSyntax-highlighted code on a lit, bevelled 3D slab that turns through space and settles face-on.GitHub Dark, Dimmed, One Dark, Draculacode-slab-3d.md · json
Desktop notification bannerA macOS-style notification banner with an app icon that springs in at the top right and tucks back out.Light, Dark, Stack, Huge (phone)desktop-notification-banner.md · json
Desktop terminal windowAn editable classic desktop terminal window (macOS Terminal-style) that types and runs a short session.Basic (light), Pro (dark), Green retro, Amber retrodesktop-terminal-window.md · json
Exploded UI panelsAn app screen's layers float apart in 3D around an orbiting camera, then slide home into the finished screen.Orbit and collapse, Isometric stack, Phone layersexploded-ui-panels.md · json
Phone message threadA messaging-app conversation (iMessage-style bubbles) that plays out bubble by bubble for hooks, ads and stories.Two people, Typing (dark), Group chat, Photo, voice memo, link preview, Reactions + read receipt, Huge textsphone-message-thread.md · json
Team chat channelMessages, unfurls, threads and reactions landing in a Slack-style channel for B2B launches and social proof.Conversation, Link unfurl, Thread, Reactions, Huge typeteam-chat-channel.md · json
Terminal command runA command typed and run in a modern GPU terminal (Ghostty-style), with output, a progress render or a git log.Command and output, Render with progress, Git log, Huge typeterminal-command-run.md · json

Workflow

  1. Match the request to a template and one or two presets from the table. Say which and why in one line, and show the person the preset names with their one-line looks before building anything.
  2. Read the reference for that template, and load its JSON. Take values from the JSON; do not re-guess them.
  3. Build it the way the reference's "Build it" section says, starting from the named HyperFrames blocks or varg components where it names them. Drive every value from timeline time so any frame renders the same on a re-render.
  4. Fit it to the brand: swap in the person's colors and fonts only where the reference allows, then re-check contrast and safe zones with the numbers in "Timing and numbers".
  5. Check a draft: render stills at the key moments (entrance, hold, exit), go through "Mistakes that look amateur", and fix before the final render.

Rules that hold for every template

  1. Rebuild the UI as HTML, don't screen-record: clean surface, deterministic timing.
  2. Hold readable UI text clamp(0.6 + chars/15, 1.0, 6.0) s (238 wpm reading, 20 cps subtitle cap).
  3. Typing 12–20 cps with seeded uneven gaps (33–200 ms, longer at spaces and before flags); 30 cps for long commands; agent/system text streams 40–80 cps in word bursts.
  4. After a command hold 3 s (short output) / 5 s (long); terminal demos under 30 s; speed up long output instead of cutting.
  5. Entrances ease-out, <300 ms for UI parts, transform + opacity only, from scale ≥0.93 (chat bubbles 0.75 from the sender's corner), never from 0.
  6. Camera moves ~1 s before the action; zoom clamped to UI bounds; settle before text must be read.
  7. Desktop UIs drawn at 1.25–1.5x; readable text ≥2.2% of frame height (terminal/code 2.6–3.3%); show only the window.
  8. Theme colors from MIT theme sources; never mix palettes (Dracula, GitHub Dark/Dimmed, One Dark, VS Code 2026).
  9. OFL stand-ins: Inter for SF Pro, Lato for Slack, JetBrains Mono for SF Mono, VT323 for retro.
  10. No platform logos; invented people and workspaces; no invented quotes as real.

Built from

Charm VHS source + practice guide; Emil Kowalski (@emilkowalski); Adam Pietrasiak (@pie6k, Screen Studio); Netflix Timed Text + Brysbaert; VS Code/Primer/Dracula/One Dark theme sources; Apple HIG; Samuel Kraft; Thorsten Ball ("How to Demo"); David Sheldon-Hicks (Territory Studio).

Every rule in the references cites its source by number. The full list, with links and what was taken from each, is in references/sources.md. Values marked as house defaults in a reference are our own starting points where no expert published a number.

Related skills

  • varg-video-edit-library: UI walkthroughs, cursors, device mockups, code and terminal edits.
  • varg-imessage-video (code2video-templates): a ready iMessage renderer.
  • HyperFrames registry (npx hyperframes catalog --query "…"): about 450 ready-made blocks; the references name the ones to start from.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.