Overview

Why does this feature matter?
No longer does a task like monitoring domain performance require the use of custom API scripts or dashboards; rather, Mailgun's MCP empowers you to accomplish such a task simply by asking for it to be done.

With Mailgun’s open-source Model Context Protocol (MCP) server, you can harness the power of AI to seamlessly query email performance metrics using nothing more than natural language. Simply put, we've enabled AI assistants like Claude to interact with Mailgun's APIs.

Feature Availability:
Mailgun's MCP server is available on all plans.

How do I get started?

Our MCP server implementation is available on GitHub with the steps to get started outlined in the README. For more technical information about Mailgun's MCP server, please refer to this blog post.

What can it do?

With Mailgun MCP, you can:

• Send emails
• Gain quick insights without needing control panel dashboards
• Obtain all sorts of email performance metrics without a single line of code
• Pull delivery statistics, bounce logs, and domain settings
• Generate reports and visualizations of your email data
• Improve workflow efficiency with AI-driven queries
• Validate email addresses

Read-only access (with one exception):
Only read actions can be performed using Mailgun's MCP; the one exception is the ability to send (i.e. create) emails. The reasoning for this restriction is to allow you the freedom to fully explore your account's data without the risk of accidentally impacting your production environment.

What does it look like in action?

Below are two screenshots showing an incredible amount of highly detailed and nuanced deliverability data: prompted using nothing more than a single sentence written in natural language. 

• "How many messages have been sent on the example.com domain over the last 7 days?" 
  • This query sought quantitative data on message volume for a specific domain within a defined timeframe.
• "Show me failed messages for the example.com domain over the last 7 days?" 
  • A technical investigation into delivery failures, requesting granular data on message rejection patterns.
• "Can you create for me a comprehensive deliverability report for example.com, please include charts and graphs to visualize the results?"
  • A request for advanced analytics visualization of sending performance metrics.
• "Can you try again. The metrics should track delayed_bounce_count and soft_bounces_count"
  • This additional context provided guidance on specific metric parameters to include in the API calls to retrieve more accurate bounce data.
• "Can you show me the performance of the example.com domain over the last 7 days?"
  • This displayed a broader performance analysis beyond just bounces, asking for comprehensive domain performance metrics.
• "Can you show my soft bounces, hard bounces, and suppressed bounces for example.com over the last 7 days?"
  • This request provided a breakdown of bounce metrics by category.
• "What caused the spike in failures for my domain example.com on July 18 2025?"
• "Can you tell me how my domain's example.com deliverability has performed over the last 3 days?"
• "Can you create some charts and graphs of the analysis?"
• "How many subaccounts do I have?"

Need Support?

Our Support Team here at Sinch Mailgun is happy to help! Reach out to us in the Support page of your Mailgun Control Panel, and we'll be with you shortly!