Contact Us

If you still have questions or prefer to get help directly from an agent, please submit a request.
We’ll get back to you as soon as possible.

Please fill out the contact form below and we will reply as soon as possible.

  • Contact Us
  • Home
  • mxHERO AI

Installing mxMCP

mxHERO's MCP Server for powerful AI Email Search

Updated at May 12th, 2025

Contact Us

If you still have questions or prefer to get help directly from an agent, please submit a request.
We’ll get back to you as soon as possible.

Please fill out the contact form below and we will reply as soon as possible.

  • General Information
  • Pre-deployment
  • Forwarding Address (Archiving Address)
  • Email Flow (Fusion)
  • Mailbox (Email Sync)
  • Storage
    Box Service Account OneDrive for Organizations
  • Mail2Sign
    Box Sign DocuSign
  • Login
  • mxHERO AI
  • Labs
  • Basics
  • Mail2Cloud Dashboard
  • Auto Filling
  • Drag And Drop
  • MXHERO App
+ More

Table of Contents

Introduction Why mxMCP 🔑 Key Benefits of mxMCP 🎥 Video Demo 🎁 Getting mxMCP Demo Accounts Testing with the Demo Account Using mxMCP - best practices 🏛️ Architecture Contact 📚 Further Reading

Introduction

Email is the largest untapped data source in the enterprise—and the most sensitive. mxHERO Mail2Cloud bridges this gap by automatically capturing email content and attachments into AI ready vector storage in real time, securing it, with compliance tagging, and metadata enrichment. This positions organizations to responsibly leverage AI for search, summarization, and operational insights—while reducing risk.
 

Read about the challenges of applying AI to Email Data. Why Organizations Are Failing to Fully Use AI for Email

Why mxMCP

mxMCP connects selected email data with your AI Tools

mxMCP is a Model Context Protocol (MCP) Server. MCP is a framework designed to standardize the way models interact with various data sources and services. MCP is used to facilitate seamless integration to emails captured by mxHERO Mail2Cloud. Mail2Cloud is designed to selectively capture emails from one or more accounts.

🔑 Key Benefits of mxMCP

mxMCP is developed by mxHERO, an award-winning email technology firm. mxMCP incorporates decades of email domain knowledge and experience into a high performance data source for your AI applications.

  • Multi-Account: Allows search of emails across email accounts. E.g., sales team communications with leads
The volume of email is one of AI's greatest challenges.
  • Security: User's can only view email that they have authenticated access
  • Scalability: High performance search across large email stores
    • search across hundreds of thousands and even millions of emails
  • capture email communications in real-time or in archives
  • Source transparence: Ensure accuracy of AI responses, with links to all source (email & attachments).
  • High Performance: 
    • Message de-duplication
    • image OCR
    • metadata filtering
    • plus many more techniques ensuring optimal AI performance restricted by LLM model context windows.

🎥 Video Demo

Video Demonstration of mxMCP installed in Claude for Desktop.

 

🎁 Getting mxMCP

mxMCP is an open source project available for Go and Python.

  • mxMCP Go: https://github.com/mxaiorg/mxmcp
  • mxMCP Python: coming soon

You will need an mxHERO AI search account to use mxMCP. You can get a demo account at: https://lab4-api.mxhero.com/demo_signup (see notes on the demo account below). For a production account get in touch with your mxHERO representative or contact form.

Demo Accounts

You can get a demo account to test mxMCP. The demo account comes pre-loaded with email. These emails are combination of mock communications and newsletters. The messages can be reviewed at the following link: Demo Emails.

Get your demo account here: https://lab4-api.mxhero.com/demo_signup

To use the demo account read this article.

 

Testing with the Demo Account

For a helpful guide on how to test the demo account read more here.

 

 

Using mxMCP - best practices

For best performance queries should specifically scope the set of emails that need to be searched. When possible, imply a date range. Even a broad date range like, “last summer” is better than no date range at all.

For example:

🤔 Okay: 
Get the email from Ralph about Project X

👏 Better: 
Get the emails from Ralph to Alex sent last June about Project X

The second query reduces the set of emails to those sent to Alex and, importantly, includes a date range.

Without a date range, mxHERO's AI search will work its way back in time in larger and larger intervals. This takes more time to execute and may result in less accurate results.

 

🏛️ Architecture

mxHERO's Multi-Account email search MCP is architected for high-performance and data security. Security is ensured at ingestion. By leveraging mxHERO Mail2Cloud, email is collected based on specific criteria and stored to isolated data tenants. The User’s AI client only has authenticated permission to a specific data tenant. As an example, this model ensures that emails selected for the human resource team can remain only accessible to the HR team. User can not attempt to prompt “jailbreak” to content they are not authorized. Importantly, other MCP tools (potentially malicious), also do not have access to content to which the user does not have authority.

 

Contact

You can get in touch with mxHERO via our contact webform

📚 Further Reading

  • Why Organizations Are Failing to Fully Use AI for Email
unnamed nameless

Was this article helpful?

Yes
No
Give feedback about this article

Related Articles

  • Demo Account for Mail2Cloud Advanced AI Testing
astors-3-tiny.png

Company

Home

About

News

Blog

Security & Trust

Terms of Service

Privacy Policy

Service Status

Partners

Products & Papers

Data Sheets

Products

FAQ

Roadmap

Contact

Sales

Support

Videos

Japan

Report Incident

Contact Us!
  • LinkedIn

contact@mxhero.com

3x Astors Homeland Security Platinum Award for Best Email Security Solution

© 2022 mxHERO Inc.

BoxElite.png

USA

100 Pine St., Suite 1250

San Francisco, CA 94111

USA

​

Japan

105-0003

DLX Building 9F

Nishi-Shinbashi 1-13-1

Minato-ku, Tokyo

Japan

Definition by Author

0
0
Expand