Beat Sync Montage logo

Beat Sync Montage

CommunityPopular
0xsline
beat-sync-montage

Plan and build a finished beat-synced montage where cut placement serves the content, not just the metronome. Use for 卡点混剪, 卡点剪辑, 踩点视频, beat-sync montage, rhythm edit, 节奏剪辑, 音乐驱动剪辑, or when the user wants cuts that land on the beat.

Overview

Publisher0xsline
RepositoryOpenChatCut
Skill namebeat-sync-montage
Stars
1.9K
Forks
277
Bundled files
2
LicenseAGPL-3.0
Links
  • 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.

  • 2 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by 0xsline on GitHub. Read the source before you install it.

Installation

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

1

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.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
  4. 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.
2

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/0xsline/OpenChatCut.git /tmp/OpenChatCut
mkdir -p .claude/skills
cp -r /tmp/OpenChatCut/src/agent/skills/beat-sync-montage .claude/skills/beat-sync-montage
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Beat Sync Montage 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 Beat Sync Montage 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 Beat Sync Montage 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.

Beat Sync Montage

Use this workflow when the edit should be driven by music. The failure mode to avoid is a metronome edit: every cut technically lands on a beat, and the result still feels mechanical, tiring, or unrelated to what is on screen.

This workflow decides what to cut and why. It does not re-explain the music tools — when you reach an execution step, follow music-intelligence for analyze_music, music_edit_plan, and sync_cuts_to_music. Load it there.

This is a OpenChatCut-native workflow. Use the current project, source assets, asset-frame inspection, AV/script context, and OpenChatCut editing tools.

When to switch workflows

  • Beat-synced photos on a locked, empty track → music-intelligence handles this directly with music_image_plan. This workflow is for video montage.
  • One long source cut to speech rhythm → long-video-to-shorts.
  • Many clips packaged as highlights without music driving the cut → multi-clips-to-reels.
  • N distinct beat-synced variants from one pool → batch-montage-variants, using this workflow for each cut.

Workflow

  1. Confirm the track is on the timeline and identified. Read the music first: BPM, meter, sections, and energy shape. Use music-intelligenceanalyze_music or inspect_music; never request the CLAP embedding.
  2. Judge whether the material actually suits a beat edit. See references/shot-fitness.md. Footage with no clear motion onset, or whose motion contradicts the music's energy, will not improve by cutting on every beat. Say so instead of forcing it.
  3. Choose the density arc before choosing individual cut points. See references/beat-density.md. Density is a property of the whole piece, not a per-cut decision: a verse and a chorus should not be cut the same way.
  4. Assign content to the music's structure. Map sections to narrative roles — intro to the hook, verse to context, build to escalation, drop/chorus to payoff. Cut density follows that map.
  5. Pick the anchor points first: 3–6 moments where a cut must hit a downbeat or a section boundary because the content has real impact there. See references/beat-density.md. Everything else is filler rhythm between anchors.
  6. Fill between anchors at the density chosen for that section. Do not fill uniformly — uniform filling is what makes an edit feel mechanical.
  7. Reserve breathing room. See references/beat-density.md. A montage that never stops cutting exhausts the viewer and hides the anchors you built in step 5.
  8. Check motion continuity across cuts. See references/shot-fitness.md. Consecutive cuts should not fight each other in screen direction or motion axis unless the collision is deliberate.
  9. Plan with music_edit_plan, show the bounded summary and any cap warning, and get acceptance. Then apply with sync_cuts_to_music and the returned analysisRef. Follow music-intelligence for both.
  10. QA by watching, not by counting. See the rules below. A cut list that is 100% on-beat can still be a bad edit.

Plan Format

  • Track, BPM, meter, section map with timecodes
  • Density arc per section (sparse / medium / dense / rest)
  • Anchor points: frame, musical event, and what lands there
  • Shot order with the reason each shot sits where it does
  • Breathing points: where the edit deliberately stops cutting
  • Risks: weak motion onsets, motion-direction collisions, sections that outrun the material

Rules

  • Anchors over uniformity. Three or four meaningful downbeat hits beat forty evenly spaced ones.
  • Density follows the music, not the maximum. Dense cutting is a choice for a section, not a default for the piece.
  • Never cut a strong shot short to satisfy a beat. If a moment needs four beats, give it four beats and adjust around it.
  • Protect the hook. The first downbeat hit should land within the first three seconds. A beat edit that eases in has wasted its advantage.
  • Breathe. At least one held shot per 15 seconds unless the brief explicitly asks for relentless pace.
  • Motion must be legible. A cut that lands on a beat but lands mid-action with no visible onset reads as an error, not as rhythm.
  • The cut serves the content. If a section has nothing worth cutting, hold the shot and let the music carry it.
  • Do not use captions to invent claims the footage does not support.
  • Report honestly when the material does not support the requested density.

Bundled files

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

Frequently asked questions

What does the Beat Sync Montage AI skill do?

Plan and build a finished beat-synced montage where cut placement serves the content, not just the metronome. Use for 卡点混剪, 卡点剪辑, 踩点视频, beat-sync montage, rhythm edit, 节奏剪辑, 音乐驱动剪辑, or when the user wants cuts that land on the beat.

Why use Beat Sync Montage on TypingMind?

Because you install it once and use it with any model. Beat Sync Montage 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 Beat Sync Montage in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/0xsline/OpenChatCut/tree/main/src/agent/skills/beat-sync-montage. 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 Beat Sync Montage?

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 Beat Sync Montage?

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

Is the Beat Sync Montage AI skill free?

Yes. It is published on GitHub by 0xsline under the AGPL-3.0 license. 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

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇