Xeams MCP Server

Xeams MCP Server is a secure bridge that connects an AI assistant to your Xeams Email Server using the Model Context Protocol. There are two typical methods of using this server:

  1. You're using an AI assistant, such as Claude Desktop, Cursor AI, or LibreChat and need to integrate with Xeams.
  2. You're building an AI agent and need to search, send, or validate emails.

Features Exposed by Xeams MCP Server

  • Address Validation - Validates if an email address is correct. This validation is more than a simple syntax check. It confirms the domain name has an MX record and the user's address exists in the recipient's server.
  • Check Email Status - Confirms the status of a previously sent email. In this case, it will report the status of the message, the IP address of the receiving SMTP server, and whether encryption was used.
  • Server Status - Fetches the Xeams server's status, including uptime, memory usage, disk space, and email queue counts (incoming/outgoing counts, stuck/processing queue sizes).
  • Search Emails - Searches for emails processed by Xeams matching a search string, optionally scoped to a number of days back and a profile ID.
  • Fetch Raw Email - Once desired emails are fetched using the searching mechanism, you can then pull the entire email content.
  • Fetch Email Body or Attachments - Fetch just the body of the email or attachments within an email.

Prerequisites

  • You must be using the Enterprise Edition of Xeams with at least 20 users.
  • You must create an API permission in Xeams before using this MCP server. Click Home > Plugins, and click the Manage button next to Xeams API. Click the Add API Key button and create a new key with the MCP Server checked for the Permissions field. Save the generated key and secret in a safe place.

Getting Started

Download the source code for this server from GitHub. Refer to the ReadMe file for installation and usage instructions.

Practical Use Cases & Example Scenarios

Scenario 1: Automated Deliverability & Email Troubleshooting

Context: A user sent a critical invoice or notification email to a client, but the client claims they never received it.

  • Prompt to AI: Check if the email sent to john@example.com earlier today went through successfully.
  • Workflow using Xeams MCP:
    1. The AI calls Check Email Status to pull delivery logs for john@example.com.
    2. It checks the response status (e.g., delivered vs. queued/bounced), retrieves the remote SMTP server IP, and confirms whether TLS encryption was used.
    3. The AI reports back to the user: "The email was successfully delivered to john@example.com today at 10:14 AM via TLS (MX IP: 192.0.2.10). It was accepted by the destination server, so it may be in their spam or quarantine folder."

Scenario 2: Smart Customer Support & Email Search

Context: A user sent a critical invoice or notification email to a client, but the client claims they never received it.

  • Prompt to AI: Find all emails received from @acme.com in the last 7 days regarding license updates and show me what attachments were included.
  • Workflow using Xeams MCP:
    1. The AI calls Search Emails with query: "license" and days: 7.
    2. Upon finding the relevant message thread, it invokes Fetch Raw Email or Fetch Email Body or Attachments to retrieve the exact attachments.
    3. The AI summarizes the ticket history and lists the attached files directly in the chat interface.
Scenario 3: Pre-flight Recipient Validation in Autonomous AI Workflows

Context: An AI sales or support agent is preparing to send automated bulk or single transactional emails.

  • Prompt to AI: Before sending out the monthly renewal notices to list X, verify that all recipient email addresses are valid.
  • Workflow using Xeams MCP:
    1. The AI iterates over the recipient list and calls Address Validation for each address.
    2. Xeams performs MX verification and recipient verification against the target mailbox server.
    3. The AI filters out bounced or dead mailboxes before sending, protecting the domain's sender reputation.
Scenario 4: Automated IT Infrastructure Operations & Health Monitoring

Context: Give me a status update on our Xeams mail server.

  • Prompt to AI: Before sending out the monthly renewal notices to list X, verify that all recipient email addresses are valid.
  • Workflow using Xeams MCP:
    1. The AI calls Server Status.
    2. It reviews disk usage, uptime, memory, and queue sizes
    3. The AI replies: "Xeams is running normally (Uptime: 14 days, Memory: 32% used). However, there are currently 12 messages in the stuck outgoing queue that require review."