Connect MultiMail to Continue for AI Email in VS Code

Add email capabilities to Continue's open-source AI assistant through MCP. Send, read, and manage email without leaving VS Code.


Continue is an open-source AI code assistant that brings flexible, model-agnostic AI capabilities to VS Code and JetBrains. With support for local models, custom context providers, and MCP servers, Continue gives you full control over your AI-assisted development workflow. Adding MultiMail via MCP lets Continue handle email operations alongside code tasks.

Continue supports both stdio and SSE transport for MCP servers. The MultiMail MCP server uses stdio, configured in Continue's config.json file at ~/.continue/config.json. Once connected, MultiMail tools appear as available actions Continue can invoke during conversations.

This integration is especially valuable for teams who want an open-source AI assistant that can read bug reports, send deployment notifications, and manage correspondence — all without relying on proprietary tools or leaving the editor.

Get started

1

Get your MultiMail API key

Sign up at multimail.dev and create an API key from your dashboard. This key authenticates all MCP server requests.

2

Install the Continue extension

Install Continue from the VS Code marketplace if you haven't already. Open the Extensions panel (Cmd+Shift+X on macOS) and search for 'Continue' to find and install it.

3

Configure Continue's MCP settings

Open Continue's config file at ~/.continue/config.json and add MultiMail to the experimental MCP servers section.

json
{
  "experimental": {
    "modelContextProtocolServers": [
      {
        "transport": {
          "type": "stdio",
          "command": "npx",
          "args": ["-y", "@multimail/mcp-server"],
          "env": {
            "MULTIMAIL_API_KEY": "mm_live_your_key_here"
          }
        }
      }
    ]
  }
}
4

Create an agent mailbox

In the MultiMail dashboard, create a mailbox for Continue to use. Start with gated_send oversight mode so outbound emails require your approval before delivery.

5

Reload and verify

Reload the VS Code window (Cmd+Shift+P > Reload Window) to pick up the config changes. Ask Continue to list your mailboxes to confirm the MCP connection is working.


Available MCP tools

ToolDescriptionExample
send_emailSend an email from your agent mailbox to any recipient.Send a deployment summary email to the team after shipping a feature.
reply_emailReply to a received email, maintaining the thread context.Reply to a client's question about API usage with code examples.
check_inboxCheck a mailbox for new or unread messages.Check for new support tickets before starting your coding session.
read_emailRead the full content of a specific email including headers and body.Read a detailed bug report to understand reproduction steps.
create_mailboxCreate a new agent mailbox on your MultiMail account.Create a dedicated mailbox for a new open-source project.
list_mailboxesList all mailboxes on your MultiMail account.View available mailboxes to choose the right sender address.
get_threadRetrieve an entire email thread to see the full conversation history.Review a full support thread to understand context before replying.
search_contactsSearch your contact list by name, email, or tags.Find all contacts tagged 'contributor' to send a project update.
add_contactAdd a new contact to your MultiMail contact list.Save a new collaborator's email after they submit a pull request.
tag_emailTag or categorize an email for organization and filtering.Tag feature request emails with the relevant milestone number.
list_pendingList all emails currently awaiting human approval.Review queued outbound emails before approving them in batch.
decide_emailApprove or reject a pending email in the oversight queue.Approve a changelog email after reviewing its content and recipients.

Usage examples

Continue MCP configuration
json
{
  "experimental": {
    "modelContextProtocolServers": [
      {
        "transport": {
          "type": "stdio",
          "command": "npx",
          "args": ["-y", "@multimail/mcp-server"],
          "env": {
            "MULTIMAIL_API_KEY": "mm_live_your_key_here"
          }
        }
      }
    ]
  }
}

Add this to ~/.continue/config.json to connect MultiMail to Continue.

Bug report to fix workflow
text
Check my MultiMail inbox for emails tagged 'bug'. Read the latest one, find the relevant code in our project, fix the issue, and reply to the reporter with an explanation of the fix.

Ask Continue to read a bug report email, investigate, fix the issue, and reply.

Code review email summary
text
Review the git diff for the last 3 commits. Write a non-technical summary of what changed and why, then send it to [email protected] from my MultiMail mailbox.

Generate a summary of recent code changes and email it to stakeholders.

Multiple MCP servers with Continue
json
{
  "experimental": {
    "modelContextProtocolServers": [
      {
        "transport": {
          "type": "stdio",
          "command": "npx",
          "args": ["-y", "@multimail/mcp-server"],
          "env": {
            "MULTIMAIL_API_KEY": "mm_live_your_key_here"
          }
        }
      },
      {
        "transport": {
          "type": "stdio",
          "command": "npx",
          "args": ["-y", "@other/mcp-server"]
        }
      }
    ]
  }
}

Use MultiMail alongside other MCP servers in Continue.


Best practices

Continue uses an experimental MCP flag

MCP support in Continue is under the 'experimental' key in config.json. Check Continue's documentation for any updates that may move MCP configuration to a stable location in future releases.

Use Continue with local models for privacy

Continue supports local LLMs via Ollama or LM Studio. Combine a local model with MultiMail's MCP server to keep your AI conversations private while still having email capabilities. Your emails are handled by MultiMail's API, but your prompts stay local.

Leverage custom context providers

Continue's custom context providers can pull in project documentation, issue trackers, or other data alongside MultiMail's email tools. This lets you compose emails that reference up-to-date project context automatically.

Works in JetBrains too

Continue also supports JetBrains IDEs. The same MCP configuration in ~/.continue/config.json works across both VS Code and JetBrains, so you get MultiMail email tools in whichever editor you prefer.


Common questions

Does Continue support MCP servers natively?
Yes, Continue has experimental MCP support for both stdio and SSE transport. The MultiMail MCP server uses stdio transport, which Continue handles through its experimental.modelContextProtocolServers configuration. This feature is actively maintained and improving with each Continue release.
Can I use Continue with a local model and MultiMail?
Absolutely. Continue supports local models through Ollama, LM Studio, and other providers. You can pair a locally-running model with MultiMail's MCP server so your AI assistant runs privately while still having access to email tools. The MCP server communicates with MultiMail's API, but your prompts and conversations stay on your machine.
Does the Continue config work in JetBrains IDEs too?
Yes. Continue uses the same ~/.continue/config.json configuration file regardless of whether you're running it in VS Code or a JetBrains IDE. Add the MultiMail MCP server config once and it works in both editors.
How do I update the MCP server?
The npx command in the configuration always fetches the latest version of @multimail/mcp-server from npm. To pin a specific version, replace "-y" with "-y" and specify the version like "@multimail/[email protected]" in the args array. This ensures consistent behavior across your team.
What happens if Continue can't connect to the MCP server?
If the MCP server fails to start, Continue will show an error in its output panel. Check that Node.js is installed and npx is available in your PATH. You can also verify the connection by running 'npx -y @multimail/mcp-server' in your terminal to see if it starts successfully.

More MCP clients

The only agent email with a verifiable sender

Email infrastructure built for AI agents. Verifiable identity, graduated oversight, and a 38-tool MCP server. Formally verified in Lean 4.