Sendook MCP Server

MCP (Model Context Protocol) server for Sendook - an AI email communication platform that enables AI agents to send and receive emails programmatically.

Features

• Inbox Management: Create, list, and delete managed email inboxes

• Email Sending: Send emails with HTML/text content, attachments, and CC/BCC

• Message Handling: List, retrieve, and reply to messages

• Conversation Threads: View and manage email conversation threads

• Webhooks: Configure real-time notifications for email events

Related MCP server: Sendmux Email Inbox API + Sending

Installation

# Using uvx (recommended)
uvx sendook-mcp

# Or using pip
pip install sendook-mcp

Configuration

Set your Sendook API key as an environment variable:

export SENDOOK_API_KEY="your-api-key-here"

Optional: Set a custom API URL (defaults to production):

export SENDOOK_API_URL="https://api.sendook.com"

Usage with Claude Desktop

Add to your Claude Desktop configuration (~/.config/Claude/claude_desktop_config.json on macOS/Linux or %APPDATA%\Claude\claude_desktop_config.json on Windows):

{
  "mcpServers": {
    "sendook": {
      "command": "uvx",
      "args": ["sendook-mcp"],
      "env": {
        "SENDOOK_API_KEY": "your-api-key-here"
      }
    }
  }
}

Or if installed via pip:

{
  "mcpServers": {
    "sendook": {
      "command": "sendook-mcp",
      "env": {
        "SENDOOK_API_KEY": "your-api-key-here"
      }
    }
  }
}

Available Tools

Inbox Management

Message Operations

Thread Operations

Webhook Management

Examples

Send an Email

Use sendook_send_message with:
- inbox_id: "ibox_01J3ZKZ0BRQ9SSJK1GRSCX4N4Z"
- to: ["customer@example.com"]
- subject: "Your Order Confirmation"
- text: "Thank you for your order!"

Create a Support Inbox

Use sendook_create_inbox with:
- name: "Customer Support Bot"
- email: "support@yourdomain.com"

Set Up Webhook for Incoming Emails

Use sendook_create_webhook with:
- url: "https://your-app.com/webhooks/sendook"
- events: ["message.received", "message.delivered"]

Response Formats

Most read operations support two response formats:

• markdown (default): Human-readable formatted output

• json: Machine-readable structured data

Set response_format: "json" for programmatic processing.

Error Handling

The server provides actionable error messages:

• 401: Authentication failed - check your API key

• 403: Permission denied - verify API key permissions

• 404: Resource not found - check the ID

• 429: Rate limit exceeded - wait and retry

Development

# Install dev dependencies
uv sync --extra dev

# Run linting
uv run ruff check .

# Run tests
uv run pytest

License

MIT