home / mcp / vavicky mcp server
Provides a complete MCP server for the VoiceAI VAVicky AiAgency platform, handling users, tokens, assistants, Twilio, and communications via MCP.
Configuration
View docs{
"mcpServers": {
"aiagency-now-mcp-voiceai-whitelabel": {
"url": "https://backend.vavicky.com/mcp",
"headers": {
"VAVICKY_API_KEY": "YOUR_API_KEY"
}
}
}
}You deploy and run the Vavicky MCP Server to manage VoiceAI assistants, tokens, communications, and Twilio integration via MCP. It provides both a remote API endpoint and a local, server-based runtime you can run on your infrastructure, enabling dynamic configuration, monitoring, and control of your MCP-enabled workflows.
You can access the MCP server through a public HTTP URL to connect your MCP client or use a local stdio server for on-premises deployments. The HTTP endpoint serves as the remote MCP connection, while the local runtime allows you to run the MCP server directly on your machine or server. Use whichever method fits your deployment strategy, then use the available tools to manage users, tokens, assistants, Twilio integration, and communications.
Prerequisites: ensure you have Node.js and npm installed on your system.
# 1. Create a project directory
mkdir vavicky-mcp-server
cd vavicky-mcp-server
# 2. Initialize with npm and install required packages
npm init -y
npm install @modelcontextprotocol/sdk node-fetch
# 3. Create the server entry point (server.js) and add your runtime code here
# 4. Set up environment variables
echo 'VAVICKY_API_KEY=your-api-key-here' > .env
# 5. Make the server executable if you plan to run it directly
chmod +x server.jsPublic MCP URL for remote access: https://backend.vavicky.com/mcp.
To update the MCP server on AWS and restart the service while viewing logs, run the following from your project root that contains the MCP code:
cd MCP-VoiceAI-WhiteLabel/ && git pull && pm2 restart 6 && pm2 logs 6
```
To exit the logs, press CTRL+C.Claude Desktop can be configured to interface with the MCP server. Add the following configuration to your Claude Desktop config file to point to your local MCP server run configuration.
"mcpServers": {
"vavicky": {
"command": "node",
"args": ["/absolute/path/to/your/vavicky-mcp-server/server.js"],
"env": {
"VAVICKY_API_KEY": "your-vavicky-api-key-here"
}
}
}API keys are passed via environment variables. All requests should include proper authentication headers. Perform input validation on all parameters and maintain up-to-date dependencies to minimize security risks.
Retrieve user data including tokens and settings.
Update White Label branding details such as name, description, domain, and color.
Configure SMTP settings for custom email notifications.
Set the OpenAI API key used by assistants.
Set the ElevenLabs API key for voice capabilities.
Set the Deepseek API key.
Set the Google Gemini API key.
Set the Open Router API key.
List all assistants.
Get basic assistant information.
Get complete details for a single assistant.
Create a new assistant with full configuration.
Update an existing assistant.
Remove an assistant.
List files associated with an assistant.
Remove a file from an assistant.
Get usage statistics for an assistant.
Get token usage across all assistants.
Retrieve the dashboard-focused assistant.
Chat with an assistant.
Connect a Twilio account.
Disconnect Twilio from the MCP server.
List Twilio numbers owned by your account.
Find available phone numbers for purchase.
Purchase a Twilio phone number.
Configure settings for a Twilio number.
Get Twilio usage statistics.
Make an individual phone call.
Place multiple calls in a single operation.
List active calls.
Cancel an ongoing call.
Send an SMS message.