Zendesk MCP Server Actor

Pricing

Rating

Developer

Actor stats

Categories

Share

A Model Context Protocol (MCP) server for Zendesk integration, running as an Apify Actor. This server provides tools for interacting with Zendesk, including ticket management, user operations, and search functionality through the MCP protocol.

Features

• Ticket Management: Create, read, update, and comment on support tickets
• User Management: View user information and details
• Search: Advanced ticket search with flexible query options
• Resources: Access Zendesk tickets, users, groups, and ticket fields
• Prompts: Templates for creating tickets, crafting responses, and analyzing data
• Pay-per-Event Pricing: Transparent pricing model with pay-per-use for each operation

Available Tools

Available Resources

• zendesk://tickets - List of all recent Zendesk tickets
• zendesk://users - List of all Zendesk users
• zendesk://ticket-fields - List of all ticket fields (default and custom)
• zendesk://groups - List of support groups in the Zendesk account

Available Prompts

• create_ticket - Template for creating a new ticket with appropriate details
• ticket_response - Template for drafting professional responses to tickets
• ticket_analysis - Template for analyzing ticket data and trends

Configuration

Required Environment Variables

Set the following environment variables when deploying the Actor:

• ZENDESK_BASE_URL - Your Zendesk subdomain URL (e.g., https://your-subdomain.zendesk.com)
• ZENDESK_EMAIL - Your Zendesk account email address
• ZENDESK_API_TOKEN - Your Zendesk API token

To get an API token, visit Admin > Channels > API in your Zendesk Support admin interface.

Optional Configuration

• SESSION_TIMEOUT_SECS - Inactivity timeout in seconds before terminating idle sessions (default: 300)

Usage

Deploy to Apify

1. Deploy this Actor to the Apify platform
2. Configure the required environment variables in the Actor input
3. Start the Actor in STANDBY mode
4. Use the provided MCP endpoint URL to connect from your MCP client

Connect from MCP Client

Once the Actor is running in STANDBY mode, add the following configuration to your MCP client (e.g., Claude Desktop, VS Code):

{
"mcpServers":{
"zendesk-mcp-server":{
"type":"http",
"url":"https://YOUR_ACTOR_STANDBY_URL/mcp",
"headers":{
"Authorization":"Bearer YOUR_APIFY_TOKEN"
}
}
}
}

Replace:

• YOUR_ACTOR_STANDBY_URL with the standby URL provided by Apify
• YOUR_APIFY_TOKEN with your Apify API token

Example Queries

Once connected to Claude or another MCP client, you can ask questions like:

• "Show me all open tickets with high priority"
• "Create a new ticket for a customer who can't log into their account"
• "Find all tickets submitted by user@example.com"
• "What's the status of ticket #12345?"
• "Add a reply to ticket #12345 explaining how to reset a password"

Development

Local Development

1. Clone the repository:

   git clone https://github.com/Yash-Kavaiya/Zendesk-MCP-Server-actor.git
   cd Zendesk-MCP-Server-actor

2. Install dependencies:

   $pip install-e.

3. Set up environment variables:

   exportZENDESK_BASE_URL="https://your-subdomain.zendesk.com"
   exportZENDESK_EMAIL="your_email@example.com"
   exportZENDESK_API_TOKEN="your_api_token"
   exportAPIFY_META_ORIGIN="STANDBY"

4. Run the server:

   $python -m src

Project Structure

Zendesk-MCP-Server-actor/
├── .actor/
│ ├── actor.json # Actor metadata
│ ├── Dockerfile # Container configuration
│ ├── input_schema.json # Input configuration schema
│ └── pay_per_event.json # Pricing configuration
├── src/
│ ├── __init__.py # Package initialization
│ ├── __main__.py # Main entry point
│ ├── const.py # Constants and tool whitelist
│ ├── models.py # Data models
│ ├── event_store.py # Event store implementation
│ ├── mcp_gateway.py # MCP gateway logic
│ └── server.py # Proxy server implementation
├── pyproject.toml # Project configuration
├── README.md # This file
└── .gitignore # Git ignore rules

Architecture

This Actor is built on the Apify Python MCP Server Template and uses the following components:

1. MCP Gateway: Proxies requests between MCP clients and the Zendesk MCP server
2. Proxy Server: Exposes a Streamable HTTP transport endpoint at /mcp
3. Event Store: Maintains session state for connection resumption
4. Charging System: Integrates with Apify's pay-per-event pricing model

Links

• Zendesk MCP Server on PyPI
• Apify Python MCP Server Template
• Model Context Protocol Documentation
• Apify Platform

License

This project is licensed under the Apache License 2.0.

Support

For issues and questions:

• GitHub Issues: Report an issue
• Apify Support: Contact Apify

Version

Current version: 0.1.0

Credits

Built with:

• mcp-zendesk - The underlying Zendesk MCP server
• Apify SDK - Apify Python SDK
• Model Context Protocol - MCP specification