Elastic Email MCP

Please note that the MCP feature is available only for Email Marketing and Email API plans.

By implementing MCP, Elastic Email allows AI agents to access and control your email operations while maintaining security and proper authentication.

A_vailable actions our MCP will be able to do for you_

• Add contact to a list
• Upload and add contacts
• Create a list
• Create a segment
• Create a campaign
• Get campaigns
• Update a campaign
• Pause a campaign
• List campaigns
• Send emails: both transactional and bulk
• Fetch contact, contacts, contact from list
• Fetch lists
• Fetch templates
• Fetch segments
• Is ready (health check, connection and readiness check)
• Delete contact from list
• Delete contact from account

Please note that not all of the listed endpoints are available for all of Elastic Email pricing plans. For example - campaigns and contacts endpoints are not available for Email API plans. Here is the full list of our available articles on the above listed features:

https://help.elasticemail.com/en/articles/4984897-contact-lists-and-segments

https://help.elasticemail.com/en/articles/5472509-how-to-send-your-first-campaign

http://help.elasticemail.com/en/articles/2300606-how-to-manage-templates

Setup

In order to connect to our MCP server you can use any AI Agent that supports MCP. The key requirement for any setup is the support of "agent mode," which allows the AI agent to interact programmatically with external services securely.

We have prepaed an example step by step setup using VS Code and Github Copilot tools.

• https://github.com you will need to create a free account in Github
• https://code.visualstudio.com/ download VS Code from this website

Next you will need to connect your Github account to VS Code

• Install and Run Vs Code

• Open extensions on the left menu and install Github Copilot and Github Copilot Chat extensions

• Sign in to Github using the account you created in the previus step

• Authorize Visual Studio Code in Github Copilot

• Allow to run VS Code in next step

• To display chat window go to View → Chat in VS Code top menu. Chat should be displayed on right side.

• Switch the chat to Agent Mode

• Create an API key in Elastic Email. Required permissions to view and modify are as follows: Account, Templates, Campaigns, Contacts, Files, Send HTTP. On top of this, at least "view" access to Access Tokens is required. Please remember to never share your API Key with unauthorized third parties!

• In VS Code top bar use the search function to find: Show and run commands

• You can write "MCP" in the search field to narrow the available options down and choose "MCP: Open user configuration".

• Paste the configuration text like below (if you already have any servers only elasticemail.mcp data will be enough)

{
"servers": {
"elasticemail.mcp": {
"url": "https://mcp.elasticemail.com",
"headers": {
"X-Auth-Token": "your_api_key"
}
}
}
}

Please note that the most important step here is to input your actual API Key with above mentioned permissions. Also please make sure that after modifying this field you have to make sure you save the changes before going forward. Without saving the changes the option to "start" the integration will not appear.

• Now you can initiate connection to MCP server: click Start above elasticemail.mcp in configuration file.

• Alternative way: Choose Extensions from left menu, choose server from MCP SERVERS INSTALLED at the bottom, right click and choose Start server

From now you can use the actions provided by Elastic Email MCP server inside chat window. These actions were outlined in the beginning of this article.

You can try your first command in order to make sure the integration works. You can input "Is MCP working" prompt as an example and if everything works you should get a confirmation as a reply.

In order to have more understanding of how this kind of integration works on Elastic Email's end, feel free to review our API Documentation too:

https://elasticemail.com/developers/api-documentation/rest-api

Tips

1. AI agents might get stuck remembering some previous instructions and retrying them. It’s useful to start a new chat session to resolve such issues. Also you might instruct Agent to refetch tools and say that there have been changes on the server to ensure it actually refetches new data.
2. The more precise the prompt is the less room for interpretation you leave to the LLM reducing hallucination related issues.
3. It’s useful to review the requests before sending them, as you might notice potential inconsistencies on how LLM translated your request to API calls.