Every dashboard action is an MCP tool your agent can call.Read the docs
AI for email

Let your AI agent operate the email system. Keep launch control.

Your agent reads live contacts, messages, flows, and delivery evidence. It prepares templates and automations, while product-enforced controls reserve customer-visible launches for you.

Full agent access on every tier, including free.

Copilot run
AI for email

Read, propose, then act

Operator prompt

Inspect the onboarding flow, explain where contacts stop, and show the evidence before proposing a change.

Read-only first task

Read

Live flow and evidence

Draft

Optimization proposed

Approval

Waiting for operator

Consequential writes need approval

[ 01 / 03 ]What it works with

Real state, not a summary of it.

Draft from live context

Contact history, flow performance, and delivery events — read through tools, not pasted into a prompt.

Operate the whole surface

The same operations the dashboard calls, reachable from the app, the API, and MCP.

Keep the decision yours

Flows start paused, and consequential actions require explicit confirmation.

[ 02 / 03 ]How it works

From a question to a change you approved.

01

Ask it to look

It reads live contacts, messages, delivery events, and flow performance. No exports, no screenshots.

02

Let it prepare

It writes the template, builds the flow paused, or assembles the segment — naming exactly what it will change.

03

Approve what touches customers

You make the launch decision. Everything short of that, it just does.

Enforced by the product, not the prompt

  • Flows created paused
  • Immutable published versions
  • Suppression checked at dispatch
  • Kill switch and daily cap

Questions people ask us

Give your agent live email context and bounded authority.

Connect your agent, let it read your real sending state, and approve the first change it proposes.