← All guides marginal.sh

Email Campaign MCP Tools for AI Agents

Updated June 2026 · Hosted email MCP server

If you're searching for email campaign MCP tools, you want your AI agent to do more than draft copy — you want it to launch real tests, measure opens and clicks, and decide what to send next. marginal exposes exactly that surface over the Model Context Protocol.

What email campaign MCP tools actually do

An MCP server gives an AI agent a set of callable tools instead of leaving it to guess at API shapes. marginal is a hosted email MCP server, so once it's connected your agent can run an entire experiment loop without you wiring up send infrastructure.

Running a subject line A/B test through an agent

A subject line A/B test is the most common reason teams reach for email campaign MCP tools. With marginal, the agent generates a handful of variants, launches a managed send that splits recipients, and waits for engagement data to flow back.

Because the sends and tracking are handled by the hosted service, your AI agent email marketing workflow stays focused on judgment — which copy to keep, which to cut — rather than on plumbing. Lift is reported within each test, so winners are clear without exporting to a spreadsheet.

How to connect marginal

marginal runs at https://marginal.sh/mcp and authenticates with a Bearer API key (marg_live_...). It works with the MCP clients developers already use, so you can drive campaigns from your editor or assistant of choice.

Supported clients include Cursor, OpenAI Codex, Claude Desktop, Claude Code, Windsurf, Cline, Continue, and Zed. The free tier covers 100 experiments per month, which is plenty for building and validating an agent workflow before scaling up.

Get started with marginal

marginal is a hosted email marketing MCP server at marginal.sh. Sign up free, create an API key, and connect https://marginal.sh/mcp from Cursor, Codex, or Claude Desktop.

Documentation · Sign up · FAQ