Ping Principal MCP
Enables AI agents to request human input through native macOS dialogs for clarifying questions and user interactions.
README
Ping Principal MCP
An MCP server that enables AI agents to request human input through native macOS dialogs. Provides a structured way for AI systems to ask clarifying questions rather than making assumptions.
Overview
This tool implements the Model Context Protocol (MCP) to facilitate human-AI interaction through system dialogs. When an AI agent encounters uncertainty, it can invoke dialog tools to gather information from the user before proceeding.
Features
- Native macOS Dialogs - No files, no fuss, just clean system dialogs
- Multiple Dialog Types - Text input, multiple choice, yes/no, info, notifications
- Zero Configuration - Works out of the box with
npx - MCP Standard - Compatible with Claude Desktop, Cursor, and other MCP clients
- Timeout Handling - Dialogs don't hang forever
- macOS Only - Uses native AppleScript for authentic system integration
Installation & Usage
Quick Start
npx ping-principal-mcp
This will start the server and wait for MCP connections.
MCP Client Configuration
Claude Desktop
Add to ~/.claude_desktop_config.json:
{
"mcpServers": {
"ping-principal": {
"command": "npx",
"args": ["ping-principal-mcp"]
}
}
}
Cursor
Add to .cursor/mcp.json:
{
"mcpServers": {
"ping-principal": {
"command": "npx",
"args": ["ping-principal-mcp"]
}
}
}
Dialog Types
Text Input
Ask open-ended questions with text responses:
await ask_human({
type: "text",
question: "What's the database connection string?",
context: "Setting up production environment",
defaultValue: "postgresql://localhost:5432/mydb"
});
Multiple Choice
Present options for the human to choose from:
await ask_human({
type: "choice",
question: "Which authentication method should I implement?",
choices: ["OAuth2", "JWT", "Session-based", "API Keys"],
context: "Building user authentication system"
});
Yes/No Confirmation
Get confirmation before taking actions:
await ask_human({
type: "confirm",
question: "Should I delete the old migration files?",
context: "Database cleanup process",
defaultAnswer: false
});
Information Display
Show information that needs acknowledgment:
await ask_human({
type: "info",
question: "Migration completed successfully! 47 records updated.",
context: "Database migration finished"
});
Notifications
Send system notifications:
await notify_human({
message: "Build completed successfully",
title: "Development Server",
subtitle: "Ready for testing",
sound: true
});
CLI Options
npx ping-principal-mcp [OPTIONS]
OPTIONS:
-h, --help Show help message
-v, --version Show version number
-t, --timeout SEC Set dialog timeout in seconds (default: 300)
-p, --port PORT Run in HTTP mode (not yet implemented)
-H, --host HOST Host for HTTP mode (not yet implemented)
Examples
Development Workflow
# Start the MCP server
npx ping-principal-mcp
# Now your AI can ask questions like:
# "Should I use TypeScript strict mode for this project?"
# "What's the API endpoint for user authentication?"
# "Should I add error handling for this edge case?"
With Custom Timeout
npx ping-principal-mcp --timeout 600 # 10 minute timeout
How It Works
- AI gets stuck → calls
ask_human()tool - Native dialog appears → clean macOS system dialog
- Human responds → types answer or makes selection
- AI continues → uses the human response to proceed
No files to manage, no polling, no complexity. Just direct human-AI communication through the OS.
Requirements
- macOS only (uses AppleScript for native dialogs)
- Node.js 18+
- MCP client (Claude Desktop, Cursor, etc.)
Dialog Timeout
Dialogs automatically timeout after 5 minutes (300 seconds) by default. You can customize this:
- Per dialog: Set
timeoutparameter in tool calls - Globally: Use
--timeoutCLI option - Result: Timed out dialogs return a cancellation response
Error Handling
- User cancels: Returns cancellation status instead of error
- Timeout: Graceful timeout with clear messaging
- AppleScript errors: Proper error reporting to AI
- Invalid input: Validation and sanitization
Development
From Source
git clone https://github.com/willwillems/ping-principal-mcp.git
cd ping-principal-mcp
npm install
npm run build
npm start
Project Structure
ping-principal-mcp/
├── src/
│ ├── index.ts # CLI and main entry
│ ├── server.ts # MCP server implementation
│ ├── dialogs.ts # Dialog manager & AppleScript
│ └── types.ts # TypeScript interfaces
├── dist/ # Compiled JavaScript
└── package.json
Building
npm run build # Compile TypeScript
npm run dev # Watch mode
Troubleshooting
"Permission denied" or dialogs not showing
- Check macOS privacy settings
- Grant terminal access to display dialogs
- Try running from Terminal.app vs integrated terminals
"Module not found" errors
- Make sure you're on macOS
- Check Node.js version (needs 18+)
- Try
npm installin the project directory
MCP client not connecting
- Verify the configuration file syntax
- Check that the file path is correct
- Restart your MCP client after configuration changes
Contributing
Contributions welcome! Please feel free to submit issues and pull requests.
License
MIT - see LICENSE file for details.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.