Skip to main content
Brew ships an MCP server that exposes the whole platform as tools: design generation, editing, audiences, sends, automations, and analytics. Connect it to the AI client you already work in. Your email ops become part of the same chat where you plan them. This recipe walks through one realistic session: draft a newsletter, QA it, test it, and send it, all from chat. The pattern extends to everything else the tool catalog covers.
Before you start. You need a Brew account and an AI client that supports plugins or remote MCP connections (ChatGPT, Claude, Grok, Cursor, Codex, VS Code, and most agent tools). A verified sending domain is also required if you plan to send for real.

​
1. Connect Your Client

Create a plugin in ChatGPT, or add the connection URL as a custom connector in Claude or Grok. For other clients (Cursor, VS Code, Codex, and most others), add it to the MCP config directly:
On first connect your browser opens: sign in with Brew and pick a brand. The token is issued and scoped to that brand, and you can revoke it anytime in Settings → MCP. Per-client steps are on Connect your client.
Ask the agent “What can you do with Brew?” as your first message. It calls get_brew_capabilities and reports the tools and guided workflows available.

​
2. Draft the Email

Describe the send the way you’d brief a designer:
“Create this week’s product update newsletter. Three sections: the new inbox previews feature, the two bug fixes from the changelog, and a customer quote. Keep it text-forward with one CTA to the changelog.”
The agent calls create_email and returns a preview image and an Open in Brew link. If the design is still generating, the agent polls get_email with the emailId and runId instead of calling create_email again, which would start a second billed design. Click the link to see the design in the app. Email is visual, so review the render, never just the agent’s description of it.

​
3. Iterate in the Same Chat

Edits go through edit_email, so revision is conversational:
“Shorten the intro to two sentences and swap the CTA button to say ‘Read the changelog’.”
Every edit creates a new version in Brew, so you can restore an earlier one if a change goes the wrong way. A rename, a new inbox subject line, or a group move on its own goes through update_email_metadata instead, which is free and creates no version.

​
4. QA Without Leaving Chat

The full pre-send routine is available as tools:
  • “Preview this across Gmail, Outlook, and iOS dark mode” runs test_email_rendering, which starts a paid rendering job and returns a previewId. The agent polls get_email_rendering with that previewId and returns a screenshot per client.
  • “Audit this before we send it” runs audit_email and returns readiness, check coverage, size metrics, and the findings to fix. A partial result is free and does not establish readiness.
  • “Send me a test” runs send_email with test: true, which delivers a one-off test to your inbox and never touches your audience.
  • “Check whether this lands in the inbox or spam” runs create_inbox_placement_test, a seed send across real providers. It returns queued or running, so ask the agent to poll get_inbox_placement_results with the emailId and testId until it completes.
  • “Validate the Newsletter subscribers audience before we send” reads the audience’s addresses with search_contacts and checks them with validate_contacts, so invalid addresses are caught before they bounce.

​
5. Send It

Ask for the real send:
“Send it to the Newsletter subscribers audience tomorrow at 9am.”
An email send is irreversible, so Brew makes the agent stop first. send_email returns a confirmation request with the exact recipient count. The agent shows it to you, and the send only goes out after you approve. Treat that pause as part of the workflow, and read the count before you say yes.

​
6. Where This Goes Next

The same session can run the rest of your email program:
  • Audiences. “Create an audience of contacts on the Pro plan who joined this quarter” (create_audience).
  • Automations. “Build a 3-step welcome flow triggered by user signup” (save_automation), then test and publish it from chat.
  • Analytics. Next week: “How did the newsletter perform against our last three sends?” (get_email_analytics with report: { kind: "sends" } for the per-send stats, or kind: "overview" for the totals, rates, and timeseries the app shows).
  • Design import. “Turn this Figma frame into an email” (import_email with source.kind set to figma), which converts one frame deterministically and for free.
  • Send control. A gradual send already in flight can be held and restarted with control_email_send, passing its sendId and an action of pause or resume.
For the full tool list see MCP tools, and for multi-step patterns see MCP workflows. If you’re wiring Brew into a coding agent that also builds your product, Use Brew with AI agents covers that side. If your agent runs shell commands instead of speaking MCP, the Brew CLI gets it the same reach.

​
Need Help?

Our team is ready to support you at every step of your journey with Brew. Choose the option that works best for you:

Search Documentation

Type in the “Ask any question” search bar at the top left to instantly find relevant documentation pages.

ChatGPT/Claude Integration

Click “Open in ChatGPT” at the top right of any page to explore it further with ChatGPT or Claude.