@bubblyphone/mcp-server

MCP server for BubblyPhone — make AI phone calls, manage voice agents, and track billing from any AI assistant.

What is BubblyPhone?

BubblyPhone is a telephony platform for AI voice agents. You get a real phone number, connect an AI model (Gemini, GPT), give it a system prompt, and it handles phone calls autonomously — booking reservations, answering support questions, qualifying leads, or anything you can describe in plain English.

This MCP server lets you control all of that from any AI assistant that supports the Model Context Protocol.

Related MCP server: @saperly/mcp

Quick Start

Claude Desktop

Add to your config (~/Library/Application Support/Claude/claude_desktop_config.json):

{
  "mcpServers": {
    "bubblyphone": {
      "command": "npx",
      "args": ["@bubblyphone/mcp-server"],
      "env": {
        "BUBBLYPHONE_API_KEY": "bp_live_sk_your_key_here"
      }
    }
  }
}

Cursor

Add to .cursor/mcp.json:

{
  "mcpServers": {
    "bubblyphone": {
      "command": "npx",
      "args": ["@bubblyphone/mcp-server"],
      "env": {
        "BUBBLYPHONE_API_KEY": "bp_live_sk_your_key_here"
      }
    }
  }
}

ChatGPT

Connect to the remote MCP server at https://agents.bubblyphone.com/mcp — authentication is handled via OAuth.

Remote (Streamable HTTP)

Connect any MCP client to https://agents.bubblyphone.com/mcp with your API key as a Bearer token.

Available Tools

Calls

Phone Numbers

Billing

Models & Rates

Account

Use Cases

• "Call this restaurant and make a reservation for Friday at 7pm"

• "Set up a 24/7 customer support line for my business"

• "Call my doctor's office and reschedule my appointment"

• "Create a phone agent that qualifies sales leads"

Get Your API Key

1. Sign up at agents.bubblyphone.com

2. Go to Dashboard > API Keys

3. Create a new API key

Development

# Install dependencies
npm install

# Build
npm run build

# Run locally (stdio)
BUBBLYPHONE_API_KEY=your_key node dist/bin/cli.js

# Run locally (HTTP)
node dist/index.js --http --port 3100

License

MIT