Clipboard to Supabase MCP Helper
A local agent that monitors your system clipboard, automatically uploads copied images to Supabase Storage, and replaces the clipboard content with the public URL for immediate sharing.
Tools
upload_clipboard_image
cleanup_old_files
README
Clipboard to Supabase MCP Helper
A local agent that monitors the system clipboard, uploads any copied image to Supabase Storage, then writes the public (or signed) URL back to the clipboard.
Features
- Zero-click image hosting: Copy an image, get a URL instantly
- Low latency: Under 800ms from copy to URL
- Cross-platform: Works on macOS, Windows, and Linux
- MCP integration: Expose clipboard image upload as an MCP endpoint
- Auto-start: Configure to run at system startup
- Efficient detection: Hash-based deduplication with low CPU usage
Prerequisites
- Node.js 18+
- Supabase account with Storage enabled
- Platform-specific dependencies:
- macOS:
pngpaste(brew install pngpaste) - Windows/Linux: Native OS clipboard access
- macOS:
Installation
- Clone the repository:
git clone https://github.com/martinbowling/clipboard-to-supabase-mcp-helper.git
cd clipboard-to-supabase-mcp-helper
- Install dependencies:
npm install
- Create a
.envfile based on.env.example:
SUPABASE_URL=https://<project>.supabase.co
SUPABASE_SERVICE_ROLE_KEY=your-service-role-key
BUCKET=media
MCP_PORT=3333
- Test your Supabase connection:
npm run test:supabase
- Build the project:
npm run build
- Install as a system service:
For macOS:
npm run install:macos
For Linux:
npm run install:linux
For Windows:
npm run install:windows
Usage
Once installed and running, the helper will:
- Monitor your clipboard for image changes (polling every 300ms)
- Upload any copied images to your Supabase bucket
- Place the public URL back in your clipboard, ready to paste
Running the Service
The clipboard helper can be run in two modes:
Stdio Mode (Default)
npm start
This runs the MCP server with StdioServerTransport, ideal for command-line usage.
HTTP Mode
npm run start:http
This runs an Express HTTP server on port 3333 (configurable) with a proper REST API endpoint.
MCP Integration
The helper exposes the following MCP endpoints:
Upload Clipboard Image
With HTTP server mode:
POST http://localhost:3333/mcp
Request body:
{
"id": "1",
"jsonrpc": "2.0",
"method": "tool",
"params": {
"name": "upload_clipboard_image",
"input": {}
}
}
Response:
{
"id": "1",
"jsonrpc": "2.0",
"result": {
"content": [
{
"type": "text",
"text": "https://your-project.supabase.co/storage/v1/object/public/media/clips/xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx.png"
}
]
}
}
Cleanup Old Files
The helper also provides an MCP endpoint to manually trigger cleanup of old files.
Request body:
{
"id": "1",
"jsonrpc": "2.0",
"method": "tool",
"params": {
"name": "cleanup_old_files",
"input": {
"days": 30
}
}
}
Response:
{
"id": "1",
"jsonrpc": "2.0",
"result": {
"content": [
{
"type": "text",
"text": "Cleanup completed: Deleted 5 files older than 30 days. Failed: 0."
}
]
}
}
The days parameter is optional. If not provided, it will use the value from the RETENTION_DAYS environment variable.
How It Works
- Change Detection: Polls clipboard every 300ms and computes SHA-1 hash of image data
- Deduplication: Only processes new or changed images based on hash comparison
- Platform Adaptation: Uses platform-specific methods to capture clipboard images
- Supabase Integration: Uploads images to your Supabase bucket with unique UUIDs
- MCP Endpoint: Exposes functionality to AI agents via Model Context Protocol
- Automatic Cleanup: Periodically removes images older than the configured retention period (default: 30 days)
Platform-Specific Notes
macOS
- Requires
pngpaste: Install withbrew install pngpaste - Uses LaunchAgents for auto-start
Windows
- Uses PowerShell's System.Windows.Forms.Clipboard for image capture
- Uses Windows Registry for auto-start
Linux
- Uses xclip (X11) or wl-paste (Wayland) for clipboard access
- Uses systemd for auto-start
Development
# Run with live reload (stdio mode)
npm run dev
# Run with live reload (HTTP mode)
npm run dev:http
# Build for production
npm run build
# Run stdio version
npm start
# Run HTTP version
npm run start:http
License
MIT
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。