Clean-Cut-MCP
Enables users to create professional React-powered videos through Claude Desktop using natural language and the Remotion framework. It provides a persistent studio environment for generating animations, managing video assets, and rendering high-quality content.
README
Clean-Cut-MCP: AI Video Generation for Claude Desktop
Transform natural language into professional videos using Claude Desktop + Remotion.
Simply describe your animation, and watch Claude create stunning React-powered videos in a persistent studio environment.
🚀 INSTALLATION (5 minutes)
Step 1: Pull Container
docker pull endlessblink/clean-cut-mcp:latest
Step 2: Run Container
docker run -d --name clean-cut-mcp -p 6970:6970 -v ./clean-cut-exports:/workspace/out -v ./clean-cut-workspace:/workspace --restart unless-stopped endlessblink/clean-cut-mcp:latest
Step 3: Configure Claude Desktop
Option A: Quick Installer (Recommended for Windows)
powershell -ExecutionPolicy Bypass -Command "Invoke-WebRequest -Uri 'https://raw.githubusercontent.com/endlessblink/clean-cut-mcp/master/install-quick.ps1' -OutFile 'install-quick.ps1'; powershell -ExecutionPolicy Bypass -File 'install-quick.ps1'"
Option B: Manual Installation
# Windows PowerShell:
powershell -ExecutionPolicy Bypass -Command "Invoke-WebRequest -Uri 'https://raw.githubusercontent.com/endlessblink/clean-cut-mcp/master/install.ps1' -OutFile 'install.ps1'; powershell -ExecutionPolicy Bypass -File 'install.ps1'"
# Linux/macOS:
curl -fsSL https://raw.githubusercontent.com/endlessblink/clean-cut-mcp/master/install.ps1 | pwsh -ExecutionPolicy Bypass -
Step 4: Restart Claude Desktop
Close Claude Desktop completely and restart it.
✅ Result: Remotion Studio at http://localhost:6970 + Claude Desktop integration ready
🆕 What's New in v2.1.0
- 🛡️ Animation Validation System: Prevents syntax errors in generated code
- 🔧 MCP Singleton Management: No more duplicate processes or conflicts
- 📝 Config Preservation: Safely preserves existing MCP servers during installation
- 🎯 Auto-Fix System: Automatically corrects common TypeScript issues
- ⚡ PowerShell Execution Bypass: No more security policy issues
🎬 Quick Demo Test
After installation, try these commands in Claude Desktop:
User: "Create a simple welcome animation with the text 'Hello AI Video Generation'"
User: "List all existing animation components"
User: "Get the Remotion Studio URL"
🎯 Use Cases
Content Creators
- YouTube Intros: Professional animated intros with your branding
- Social Media: Quick animations for Instagram, TikTok stories
- Explainer Videos: Animated sequences for complex topics
- Brand Animations: Logo reveals, product showcases
Marketers & Businesses
- Ad Creatives: Animated ads for multiple platforms
- Presentation Graphics: Data visualizations, animated charts
- Website Assets: Loading animations, interactive elements
- A/B Testing: Generate multiple animation variants quickly
Educators & Trainers
- Course Content: Visual explanations of complex concepts
- Interactive Lessons: Animated diagrams and processes
- Student Projects: Help students create professional presentations
- Training Materials: Animated guides and tutorials
🏗️ Architecture
Claude Desktop (Windows/Mac/Linux)
↓ MCP Protocol (STDIO transport)
Docker Container (clean-cut-mcp)
↓ Internal Services
Remotion Studio (Port 6970) + MCP Server + Node.js Runtime
↓ Volume Mount
./clean-cut-exports/ (Your Videos)
System Requirements
- Docker Desktop installed and running
- Claude Desktop installed
- 4GB RAM recommended for video rendering
- 2GB disk space for container and videos
🛠️ Available MCP Tools
Clean-Cut-MCP provides 18 professional animation tools:
Core Animation Tools:
create_animation- Generate animations from natural languagelist_existing_components- View available animation templatesrebuild_compositions- Refresh animation registrydelete_component- Remove animations safely
Professional Features:
generate_with_learning- AI-powered animation with learned preferencesrecord_user_correction- System learns from your feedbackupload_asset- Add images, logos, audio to animationsformat_code- Professional code formatting
🔧 Troubleshooting
Installation Issues
# Check container is running
docker ps | grep clean-cut-mcp
# Check Remotion Studio is accessible
curl http://localhost:6970
# Check Claude Desktop configuration
cat ~/.config/Claude/claude_desktop_config.json | grep clean-cut-mcp
PowerShell Security Issues
If you get execution policy errors, use the bypass commands shown in Step 3.
Port Conflicts
If port 6970 is in use:
# Use alternative port
docker run -d --name clean-cut-mcp -p 6971:6970 -v ./clean-cut-exports:/workspace/out -v ./clean-cut-workspace:/workspace --restart unless-stopped endlessblink/clean-cut-mcp:latest
📚 Documentation
- User Guide - Complete tool reference
- Troubleshooting - Common issues and solutions
- Examples - Sample animations and use cases
🤝 Community & Support
- GitHub Issues: Bug reports and feature requests
- Discussions: Share animations, ask questions
- Docker Hub: Pre-built container images
📄 License
This project uses Remotion, which has special licensing requirements:
- Free for: Individuals, organizations ≤3 employees, non-profits, evaluation
- Company license required: For-profit organizations >3 employees
- Details: Check remotion.pro for complete licensing terms
Clean-Cut-MCP container/MCP server code: MIT License Remotion dependency: Subject to Remotion's license terms
🎬 Ready to start creating? Follow the 4 installation steps above!
Questions or issues? → Check the Troubleshooting guide
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。