How to update AWS MCP server
Public Streamable HTTP URL MCP Server
https://backend.vavicky.com/mcp
Update the MCP Server
This command will update the code on AWS and restart the MCP Server and it opens the logs.
cd MCP-VoiceAI-WhiteLabel/ && git pull && pm2 restart 6 && pm2 logs 6
Exit the logs
Press CTRL+C
Vavicky MCP Server
Complete MCP server implementation for the VoiceAI VAVicky AiAgency platform.
Features
User Management
- Get user data including tokens and settings
- Update White Label details (name, description, domain, color)
- Update SMTP settings for custom email notifications
API Token Management
- Update OpenAI API Key
- Update ElevenLabs API Key
- Update Deepseek API Key
- Update Google Gemini API Key
- Update Open Router API Key
Assistant Management
- Get all assistants
- Get specific assistant information
- Create new assistants with comprehensive configuration
- Update existing assistants
- Delete assistants
- Manage assistant files
- Get usage statistics and analytics
- Chat with assistants
Twilio Integration
- Connect/disconnect Twilio accounts
- Get available phone numbers
- Purchase phone numbers
- Manage existing numbers
- Get usage statistics
Communication
- Make individual phone calls
- Make bulk phone calls
- Send SMS messages
- Cancel active calls
- Monitor calls in progress
Installation
-
Clone or create the project:
mkdir vavicky-mcp-server cd vavicky-mcp-server -
Create package.json:
npm init -y npm install @modelcontextprotocol/sdk node-fetch -
Create the server.js file with the provided code
-
Set up environment variables:
echo 'VAVICKY_API_KEY=your-api-key-here' > .env -
Make the server executable:
chmod +x server.js
Claude Desktop Configuration
Add this to your Claude Desktop config file:
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\\Claude\\claude_desktop_config.json
{
"mcpServers": {
"vavicky": {
"command": "node",
"args": ["/absolute/path/to/your/vavicky-mcp-server/server.js"],
"env": {
"VAVICKY_API_KEY": "your-vavicky-api-key-here"
}
}
}
}
Usage Examples
Once configured, you can use these tools in Claude:
Assistant Management
Claude, create a new voice assistant named 'Sales Bot' with OpenAI GPT-4 that can make phone calls and has a professional greeting.
Communication
Claude, send an SMS to +1234567890 using assistant ID 'assist_123' with the message 'Hello, this is a test message.'
Analytics
Claude, show me the usage statistics for all my assistants this month.
Phone Management
Claude, get me a list of available toll-free numbers in the US and purchase one for my business.
Tool Reference
User Management Tools
get_user- Get user data including tokens and settingsupdate_white_label- Update White Label brandingupdate_smtp- Configure email settings
Token Management Tools
update_openai_token- Set OpenAI API keyupdate_elevenlabs_token- Set ElevenLabs API keyupdate_deepseek_token- Set Deepseek API keyupdate_gemini_token- Set Gemini API keyupdate_openrouter_token- Set Open Router API key
Assistant Tools
get_assistants- List all assistantsget_assistant- Get basic assistant infoget_one_assistant- Get complete assistant infocreate_assistant- Create new assistantupdate_assistant- Update assistant settingsdelete_assistant- Remove assistantget_assistant_files- List assistant filesdelete_assistant_file- Remove assistant fileget_assistant_usage- Get usage statsget_assistants_token_usage- Get token usage across assistantsget_dashboard_assistant- Get dashboard assistantchat_with_assistant- Chat with assistant
Twilio Tools
connect_twilio- Connect Twilio accountdisconnect_twilio- Disconnect Twilioget_twilio_numbers- List phone numbersget_available_numbers- Find available numbersbuy_twilio_number- Purchase phone numberupdate_twilio_number- Configure number settingsget_twilio_usage- Get Twilio usage stats
Communication Tools
make_call- Make phone callmake_bulk_call- Make bulk callsget_calls_in_progress- List active callscancel_call- Cancel active callsend_sms- Send SMS message
Error Handling
The server includes comprehensive error handling:
- API authentication errors
- Invalid parameters
- Network connectivity issues
- API rate limiting
Development
To run in development mode with debugging:
npm run dev
Security
- API keys are passed via environment variables
- All requests include proper authentication headers
- Input validation is performed on all parameters
Support
For issues with the Vavicky platform, visit: https://github.com/Business-On-Steroids/n8n-nodes-voiceai For MCP-related issues, refer to the Anthropic MCP documentation.
License
This MCP server follows the same licensing as the original Vavicky n8n node implementation.