OpenCut: The Open-Source Video Editor With 87,000+ GitHub Stars That's Replacing CapCut for Developers
OpenCut is a free, privacy-first open-source video editor with 87K+ GitHub stars. Built with TypeScript, Next.js, and Rust/WASM, it offers CapCut-level simplicity with AI agent integration, plugin support, and no paywalls.
If you've been using CapCut for quick video edits, you've probably noticed that more and more basic features are getting paywalled. Transitions that used to be free? Now premium. Simple text animations? Subscription required. That frustration is exactly why OpenCut has exploded to over 87,000 GitHub stars — it's the open-source video editor that gives you CapCut-level simplicity without the paywalls, privacy invasions, or vendor lock-in.
But OpenCut isn't just a clone. It's a developer-first tool built with modern tech (TypeScript, Next.js, Rust/WASM) that's being rewritten from the ground up to support AI agents, plugins, headless batch rendering, and cross-platform deployment from a single codebase. Let's dive into why this project is trending and what makes it technically interesting.
What Is OpenCut and Why 87,000+ Developers Love It
OpenCut is a free, open-source video editor that works in your browser or as a native desktop app. It aims to replicate the simplicity that made CapCut popular while addressing its biggest pain points:
- Privacy: Your videos never leave your device. No cloud uploads, no data harvesting.
- Cost: Core editing features are free. No surprise paywalls for basic transitions or effects.
- Simplicity: A clean, intuitive interface that doesn't require a film degree to use.
- Open source: MIT licensed. Fork it, modify it, self-host it — it's yours.
The project is currently undergoing a major rewrite. The "classic" version (built with Next.js and TypeScript) powers the live app at opencut.app, while the new version adds a Rust-powered core for GPU-accelerated compositing, a plugin API, an MCP server for AI agent integration, and headless mode for automated batch rendering.
The Tech Stack: TypeScript Meets Rust
What makes OpenCut technically fascinating is its hybrid architecture. The web frontend is built with Next.js and TypeScript, making it approachable for most web developers. But the heavy lifting — GPU compositing, effects rendering, masks, and timeline processing — is being migrated to a Rust core that compiles to WebAssembly (WASM).
# Clone and run the classic version
git clone https://github.com/opencut-app/opencut-classic.git
cd opencut-classic
cp apps/web/.env.example apps/web/.env.local
docker compose up -d db redis serverless-redis-http
bun install
bun dev:web
# App runs at http://localhost:3000The Rust/WASM pipeline means browser-based editing can approach native performance. Effects that would normally require server-side processing — like real-time previews of complex transitions — run locally in your browser via WASM. Here's what the architecture looks like:
┌─────────────────────────────────┐
│ Next.js Frontend (TypeScript) │
│ - Timeline UI │
│ - Project Management │
│ - Plugin System │
├─────────────────────────────────┤
│ Rust Core (WASM Bindings) │
│ - GPU Compositor │
│ - Effects Engine │
│ - Mask Processing │
│ - Binary Rendering Pipeline │
├─────────────────────────────────┤
│ Platform Layer │
│ - Web (Browser WASM) │
│ - Desktop (GPUI native app) │
│ - Headless (batch/automation) │
└─────────────────────────────────┘AI Agents, Plugins, and the MCP Server
The upcoming rewrite introduces three features that make OpenCut more than just a video editor — it becomes an extensible platform:
1. Editor API
A programmatic API that lets you control the editor from code. Imagine writing a script that automatically cuts silence from a podcast recording or generates highlight reels from raw footage.
2. Plugin-First Architecture
Third-party plugins will be first-class citizens. Want to add a custom effect, integrate with a stock footage API, or build a specialized subtitle generator? The plugin system will make it possible without forking the core.
3. MCP Server for AI Agents
Perhaps the most exciting addition: an MCP (Model Context Protocol) server that lets AI coding agents interact with the editor. This means you could ask Claude, GPT, or any MCP-compatible agent to:
"Cut this video at the 2-minute mark, add a fade transition,
and export in 1080p MP4 format."The AI agent communicates with OpenCut's MCP server, which translates natural language instructions into editor operations. This is a game-changer for content creators who want AI-assisted editing without learning complex software.
4. Headless Mode and Scripting
OpenCut will support a headless mode for automation and batch rendering. Combined with a built-in scripting tab, you can create reusable edit pipelines:
// Example: Batch process a folder of videos
const videos = await getFolder('./raw-footage/');
for (const video of videos) {
await openCut.load(video);
await openCut.trimEnd('00:05:00');
await openCut.addTransition('fade', { duration: 1 });
await openCut.export({ format: 'mp4', quality: '1080p' });
}Real-World Example: Building a YouTube Shorts Pipeline
Let's say you're a developer who also creates YouTube content. You record 10-minute coding tutorials and want to extract 60-second highlight clips for YouTube Shorts. With OpenCut's upcoming scripting API and headless mode, here's how your workflow could look:
// Step 1: Load the full tutorial
await openCut.load('./tutorial-2026-08-28.mp4');
// Step 2: Use AI to find the most engaging moments
const highlights = await openCut.analyze({
detectSegments: { minEnergy: 0.7, minDuration: 30 },
// Finds segments with high visual/audio activity
});
// Step 3: Extract and format for Shorts (9:16 aspect ratio)
for (const clip of highlights) {
await openCut.createSubProject(clip);
await openCut.resize('9:16');
await openCut.addCaption('auto-generate');
await openCut.addIntro('my-short-intro.mp4');
await openCut.export({
format: 'mp4',
quality: '1080p',
filename: `short-${clip.id}.mp4`
});
}This kind of automation was previously only possible with expensive SaaS tools or complex FFmpeg scripting. OpenCut aims to make it accessible to any developer comfortable with TypeScript.
Key Benefits of OpenCut
- Completely free and open-source — MIT licensed, no subscription traps
- Privacy-first — all processing happens locally, your videos never leave your device
- Cross-platform — works in any modern browser, with native desktop apps coming
- Developer-friendly stack — TypeScript, Next.js, Rust/WASM, and Bun
- AI-ready — MCP server integration for AI agent control
- Extensible — plugin architecture for custom effects, integrations, and workflows
- Automation-capable — headless mode for batch processing and CI/CD pipelines
- Community-driven — 87,000+ GitHub stars and active Discord community
- Sponsored by Vercel and fal.ai — backed by companies that believe in open source
Frequently Asked Questions
Is OpenCut really free?
Yes. OpenCut is MIT licensed and free to use. Unlike CapCut, which has increasingly moved basic features behind a paywall, OpenCut's core editing features are completely free. The project is supported by sponsors like Vercel and fal.ai, not user subscriptions.
Can I use OpenCut offline?
Absolutely. OpenCut processes everything locally in your browser or desktop app. Your videos are never uploaded to any server. The web version works offline once loaded, and the native desktop app (built with GPUI) runs entirely on your machine.
What video formats does OpenCut support?
OpenCut supports all major video formats including MP4, MOV, WebM, and AVI for import, and exports to MP4 and WebM. The Rust/WASM core handles codec processing for efficient rendering even with large files.
How does OpenCut compare to CapCut?
OpenCut matches CapCut's ease of use for basic editing tasks — trimming, splitting, transitions, text overlays, and simple effects. Where it differs: no paywalls, no cloud uploads, open-source code you can inspect and modify, and upcoming AI agent integration that CapCut doesn't offer. CapCut still has more polished effects and templates, but OpenCut is rapidly closing that gap.
Can I contribute to OpenCut?
Yes! The project welcomes contributions, especially in timeline functionality, project management, performance optimization, bug fixes, and UI improvements. Check their Contributing Guide on GitHub. Note that the preview panel and export features are currently being refactored, so those areas are best avoided for now.
Does OpenCut work on mobile?
The web version at opencut.app works on mobile browsers, though the editing experience is optimized for desktop/laptop screens. The upcoming rewrite aims to deliver native mobile apps from the same codebase, so dedicated iOS and Android apps may be available in the future.
What is the MCP server and how do AI agents use it?
The MCP (Model Context Protocol) server is an upcoming feature that exposes OpenCut's editor functions as an API that AI coding agents can interact with. Any MCP-compatible agent (Claude Code, OpenClaw, Codex, etc.) can send instructions like "trim this clip," "add a transition," or "export at 1080p" directly to the editor. This enables AI-assisted video editing through natural language.
Want to learn the technologies powering OpenCut? Master TypeScript, Next.js, and modern web development with CoddyKit's interactive coding courses. Build real projects, practice with hands-on exercises, and level up your developer skills today.