Magic Moment

Community

win4r

magic-moment

Make a "magic moment" video — turn a creator's talking-head recording into a vertical video preserving the source narration where their Muse story replays through artifacts and brief exchanges synced to their voiceover — bubbles, typing, emoji reactions, real widgets and pages, message sounds, closing Muse lockup finisher. Use whenever a user with talking-head or selfie footage wants it turned into a shareable clip of their Muse story — "make a magic moment", "turn this video of me into...", "add the chat over my video", "retell what Muse did for me" — even if they never say the words "magic moment".

Overview

Publisher	win4r
Repository	`MuseAI-Skills`
Skill name	`magic-moment`
Stars	382
Forks	116
Bundled files	13
Links	Skill source Repository Homepage

Markdown instructions
A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.
Works with any LLM
AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.
13 bundled files
Scripts, templates, and references the model can read while it works. Files are read-only and never executed.
Open source
Published by win4r on GitHub. Read the source before you install it.

Installation

Install the Magic Moment AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

Open the app and go to Plugins → Skills.
Choose "Install from GitHub".
Paste the skill folder URL below and confirm.
https://github.com/win4r/MuseAI-Skills/tree/main/opt/hatch/skills/magic-moment
Enable the skill in any chat where you want it available.

Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills

git clone --depth 1 https://github.com/win4r/MuseAI-Skills.git /tmp/MuseAI-Skills
mkdir -p .claude/skills
cp -r /tmp/MuseAI-Skills/opt/hatch/skills/magic-moment .claude/skills/magic-moment

Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Magic Moment in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Magic Moment on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Magic Moment is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

Loaded only when it is needed
The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.
Switch models mid-chat
Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Magic Moment

The build is a guided pipeline; read the guides in order and follow them exactly — this file is only the map.

guide/story_and_canon.md — transcribe the clip, ground every beat in the real work this VM actually did, and lock the fact sheet.
guide/conversation_shape.md — choose the exchanges and artifact reveals that carry the story.
guide/visuals.md — plan and preview the visuals. Match the component to the narrated action and give the actual artifact room to be seen.
guide/screenplay.md and guide/timeline.md — assemble beats.
guide/screenplay_review.md — the review gate before any render.

The look is owned by reference/design.md (doctrine) and reference/design-system/muse-moments-kit.html (product components). Use reference/visual-storytelling.md for diagram composition and motion. Card mechanics: reference/card-spec.md; renderer contracts: reference/overlay-spec.md.

Component index

This index names every component in the kit so you can find one by capability. Grep the label in /opt/hatch/skills/magic-moment/reference/design-system/muse-moments-kit.html and copy that component's markup, which is the source of truth for the look; /opt/hatch/skills/magic-moment/reference/design.md owns how to copy, scale, and adapt it. The kit's foundations (palette, type, rules) and its thread-chrome section are renderer-drawn reference, not components to pick.

Artifacts & documents
- web artifact · finished (hero): a page Muse built, as its real screenshot.
- fullstack app · resting (icon card, figma space icon): the app Muse built, at rest before the tap.
- fullstack app · tapped (a legit full web app): the tap beat that opens that app full screen.
- file card · shipping widget replica: a document Muse produced, with the real file in the preview well.
- slides card · shipping widget replica: a slide deck Muse produced.
- user upload chip · shipping replica: a file the user uploaded, on the user side of the thread.
- completion card: work finished when no other component covers what Muse did.
- external link · shipping widget replica: a link Muse sent, unfurled with its hero and host.
Media
- generated image · renders as-is: an image Muse generated, on its own image beat.
- generated video · renders as-is: a video Muse generated, on its own video beat.
- image picker · mobile canonical (2×2 + quote-reply): the user picking one of several generated images.
- music · found and playing: a track or mix Muse found or made, playing.
- media library · organized: photos or media Muse sorted into groups.
Browser & shopping
- browser · rebuilt journey (one card, tap by default): a site Muse drove; author the browser beat's pages rather than copying this markup.
- skill task · status chip (no browser involved): a skill or connector did the work instead of the browser.
- web search · citations: an answer Muse sourced from the web, with its citations.
- shopping results · shipping widget replica: products Muse found, as the in-bubble result grid.
- purchase · order placed (canonical figma): an order Muse placed.
- browser · activity island (canonical figma): browser work in flight, as the floating glass island.
Trust & security
- connect an account · in-chat link + disclosure sheet (canonical figma): the user connecting an account Muse needs.
- secure storage · log in details (figma canonical): the log-ins already held in secure storage.
- secure storage · add log in details (figma canonical): saving a new log-in to secure storage.
- sentinel · approval card (figma canonical): Muse asking permission before a sensitive action.
- secure form fill: Muse filling a card number or password from secure storage.
Communication
- texting on your behalf · message thread (send animates): a text Muse sent for the user, in a real message thread.
- email · triage + draft ready: an inbox Muse triaged, with a draft waiting.
- email · draft in the real gmail composer (send animates): an email Muse drafted and sent from the Gmail composer.
- phone call · live + outcome: a call Muse placed, live and then its result.
- audio pill · shipping widget replica: a voice memo or audio clip in the thread.
Automation & ambient
- scheduled tasks · agent list (canonical figma): the tasks Muse runs on a schedule.
- standing watch · ambient guardian: Muse watching something over days, such as a price.
- proactive nudge · text_with_button replica: an unprompted nudge with one call to action.
- option widget · dashed choices (shipping replica): the user picking one of several written choices.
- letter · shipping widget replica: a letter Muse wrote.
- idea card · shipping widget replica: one idea Muse surfaced, with its preview and its pill. From your calls is this card's meta line, not a component of its own.
- morning brief · feed edition (canonical figma): the morning feed edition Muse published. Your day and Heads up are section headers inside it, not components of their own.
- ideas · idea rows (canonical figma): the ideas list, with one row being chosen.
Personal intelligence
- memory · recalled from weeks ago: Muse acting on something the user said weeks earlier.
- memory · person page (markdown doc, product design): a person's memory page and its open threads.
- goals · goals surface (canonical figma): the user's goals list, with a subgoal checking off.
- goals · tracking check-in (canonical figma): a check-in on a habit Muse tracks.
- calendar · conflict resolved: a calendar conflict Muse moved.
- device sync · flowing in: the user's device data syncing in.
Work & analysis
- deep research · multi-source report: a multi-source report Muse researched.
- uploaded file · analyzed: a document the user uploaded and Muse read.
- data analysis · chart from a spreadsheet: numbers Muse charted from a spreadsheet.
- long thread · digest: a long thread Muse condensed into decisions.
Meta-capabilities
- subagents · fanning out in parallel: work Muse ran across several helpers at once.
- wallet · muse wallet screen (canonical figma): the wallet screen and its payment methods.
- wallet · purchase approval (canonical figma): the user approving a purchase Muse is about to make.
- title card · chapter divider (canonical figma): a full-bleed chapter divider between story sections.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

guide/conversation_shape.md2 KB
guide/screenplay_review.md3 KB
guide/screenplay.md8 KB
guide/story_and_canon.md2 KB
guide/timeline.md2 KB
guide/visuals.md5 KB
INSTALL.md2 KB
reference/card-spec.md6 KB
reference/design-system/muse-moments-kit.html141 KB
reference/design-system/story-compositions.html4 KB
reference/design.md26 KB
reference/overlay-spec.md5 KB
reference/visual-storytelling.md3 KB

Frequently asked questions

What does the Magic Moment AI skill do?

Why use Magic Moment on TypingMind?

Because you install it once and use it with any model. Magic Moment is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Magic Moment in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/win4r/MuseAI-Skills/tree/main/opt/hatch/skills/magic-moment. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Magic Moment?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Magic Moment?

As many as you like. As long as a model supports skills, you can use Magic Moment with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Magic Moment AI skill free?

It is published on GitHub by win4r. Check the repository for licensing terms. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Graincrawl

openclaw

OrganizationPopular

Granola archive: search, sync freshness, notes, transcripts, panels, SQL counts, and Graincrawl repo work.

Openclaw Changelog Update

openclaw

OrganizationPopular

Regenerate OpenClaw release changelog sections from git history before beta, stable, or extended-stable releases.

Openclaw Ci Limits

Manage OpenClaw GitHub Actions and Blacksmith CI capacity, runner-registration budgets, fanout caps, main-push single-flight, shard sizing, hosted-runner offload, queue health, and safe…

Openclaw Debugging

Debug OpenClaw model, provider, tool-surface, code-mode, streaming, and live/Crabbox behavior by choosing the right logs, probes, and proof path before changing code, including fetching stored…

Overview

Installation

Install in TypingMind

Install in another agent

Use it in TypingMind

Works with any AI model

Skill instructions

Magic Moment

Component index

Bundled files

Frequently asked questions

Related AI Skills

Set up your own AI workspace now