kweenkl MCP Server

kweenkl MCP Server

🚀 PRE-LAUNCH: kweenkl is currently in pre-launch mode. Join our early access program to get notifications from your AI assistant!

MCP (Model Context Protocol) server that allows AI assistants to send kweenkl notifications.

What is kweenkl?

The verb "kweenkl" means to send a notification using kweenkl. When an AI can "kweenkl" you, it transforms from a conversational tool into an active assistant that can tap you on the shoulder when needed.

Pre-Launch Access

kweenkl is currently in pre-launch. During this phase:

• The service is fully functional and ready to use
• We're gathering feedback from early adopters
• Official launch coming soon with more features

Try it now with our demo webhook token: 51fa2b2d-2080-4a73-b059-7e67712d93f7

Installation

npm install

Configuration

Claude Desktop (macOS)

Edit: ~/Library/Application Support/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "kweenkl": {
      "command": "node",
      "args": ["/absolute/path/to/kweenkl-mcp-server/index.js"]
    }
  }
}

Claude Desktop (Windows)

Edit: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "kweenkl": {
      "command": "node",
      "args": ["C:\\path\\to\\kweenkl-mcp-server\\index.js"]
    }
  }
}

Other MCP Clients

Consult your MCP client's documentation for stdio server configuration. The server requires:

• Node.js executable
• Path to index.js
• Optional environment variables

Usage

Once configured, ask your AI assistant:

• "kweenkl me when you're done"
• "Send a high priority kweenkl about the deployment"
• "kweenkl the team with the summary"

Example Interactions

Basic notification:

User: "kweenkl me with 'Hello World'"
AI: I'll send that notification now.
Response: ✅ Successfully kweenkled! 📱 1 subscriber(s) notified

With priority:

User: "Send a high priority kweenkl about the server being down"
AI: I'll send a high priority notification.
Response: ✅ Successfully kweenkled! 📱 3 subscriber(s) notified

Contextual notification:

User: "Analyze this log file and kweenkl me if you find errors"
AI: I'll analyze the log file now.
[AI analyzes file and finds errors]
AI: Found 5 errors. I'll kweenkl you with the summary.
Response: ✅ Successfully kweenkled!

Getting Your Webhook Token

Option 1: Use Demo Token (Pre-Launch)

Try kweenkl immediately with our demo webhook token:

51fa2b2d-2080-4a73-b059-7e67712d93f7

This demo token lets you test the service right away. Notifications sent to this token will be delivered to the kweenkl team's test channel.

Option 2: Get Your Own Token

1. Open kweenkl iOS app (join our TestFlight for pre-launch access)
2. Open a channel
3. Tap info icon
4. Copy webhook token

Environment Variables

• KWEENKL_API_URL - API base URL (default: https://api.kweenkl.com)
• KWEENKL_DEBUG - Enable debug logging (set to 'true')

Tool Parameters

The kweenkl tool accepts the following parameters:

• webhook_token (required): Your channel's webhook token
• message (required): The notification message (max 500 chars recommended)
• title (optional): Brief title for the notification (max 50 chars recommended)
• priority (optional): Priority level - "low", "normal" (default), or "high"
• payload (optional): Custom JSON object for additional metadata

Testing

Run the test suite:

npm test

Or test manually with the MCP Inspector:

npm install -g @modelcontextprotocol/inspector
mcp-inspector node /path/to/kweenkl-mcp-server/index.js

Troubleshooting

Requirements

• Node.js 18+
• Valid kweenkl webhook token
• Internet connection

License

MIT