Audit an Email

An email audit checks whether a finished email is ready to send. Brew checks the exact HTML, subject, preview text, and sending purpose that you provide. The result combines deterministic checks with link, image, compatibility, accessibility, and copy reviews. An audit reports findings. It does not change your HTML.

Run an Audit in the App

Open the send or export dialog

Open the email that you plan to send or export. The audit uses the live HTML, subject, preview text, version, and sending purpose shown in the dialog.

Click Audit

Click Audit · 5 credits next to Test. For exports, the button sits next to Export. Brew opens the Audit tab and starts the checks.

Review the result

Start with the readiness status and score. Then review check coverage, loaded size, and findings grouped by category. Each finding explains what failed and how to fix it.

Fix and run again

Edit the email, then click Run again · 5 credits. Any change to the HTML, subject, preview text, purpose, or audit rules makes the prior result stale.

A complete audit costs 5 credits. A partial audit costs 0 credits because one or more required checks did not finish. Never treat a partial audit as ready.

Understand the Result

The completion field tells you whether every required check finished.

Completion	Readiness	Meaning
`complete`	`ready`	No blocker, error, or warning findings.
`complete`	`needs_review`	The email has warnings but no blocker or error findings.
`complete`	`not_ready`	The email has at least one blocker or error finding.
`partial`	`unknown`	A required check was unavailable, and deterministic checks found no blocker.
`partial`	`not_ready`	A required check was unavailable, and deterministic checks already found a blocker.

Complete audits score from 0 to 100. Partial audits return completion.score: null. Brew subtracts 20 points per blocker, 10 per error, and 3 per warning, then floors the score at 0. Repeated targets can apply at most three penalties per rule. Provider scores do not change the Brew score. A complete audit stays fresh for 15 minutes. A content change or a ruleset change makes it stale sooner. The audit checks:

unsubscribe content for marketing email;
empty, malformed, placeholder, and unreachable links;
missing, unreachable, non-image, and oversized images or GIFs;
HTML size and the loaded size of directly referenced assets;
subject and preview length, clarity, duplication, and placeholders;
alt text, contrast, document language, headings, and link text;
unsupported markup, scripts, forms, iframes, and client compatibility;
common email practices such as a postal address, image dimensions, and text-to-image balance.

The copy review can add warning or info findings. It cannot decide compliance, URL validity, or whether the final action is allowed. Loaded size is exact only when Brew can measure every directly referenced asset. Otherwise, the result reports a lower bound and the number of unknown assets. It never invents a total.

Choose the Sending Purpose

Use marketing for newsletters, announcements, promotions, and lifecycle email. Marketing audits check for visible unsubscribe content. The public API defaults to marketing when you omit sendingPurpose and reports that default in policy.source. A missing visible marketing unsubscribe is a warning. Direct marketing sends still insert Brew’s unsubscribe fallback before delivery. Use transactional only for a message sent because of something the recipient did, such as a receipt, verification email, or password reset. The unsubscribe check is not applicable to transactional email. In the app, Brew derives the purpose from the selected sending domain. Transactional email creation always uses transactional. A generic export without a sending-domain purpose uses the internal unknown purpose, skips unsubscribe compliance, and does not insert unsubscribe content.

Continue After a Warning

Confirmed dead links or images, an image that returns non-image content, and a known loaded size above 5 MB pause the final action. You can click Continue Anyway after reviewing these findings. Brew logs the override without storing the HTML or full URLs. Authentication failures, throttling, server errors, bot protection, timeouts, and unknown resource sizes remain warnings. Brew does not report them as dead links or images. Deterministic blockers cannot be overridden. These include missing or invalid destinations, unresolved URL merge tags, invalid image sources, a missing or overlong subject, and overlong preview text.

Run an Audit from MCP

If Brew is connected to your AI client, ask it to audit a saved design:

“Audit the welcome email before we send it.”

The agent calls audit_email with emailId. It can also pin an emailVersionId and set sendingPurpose to marketing or transactional. The tool loads that exact saved version and returns the typed, capped result.

Run an Audit from the CLI

Pass the rendered HTML and the inbox copy that you plan to use:

brew-cli emails audit \
  --file newsletter.html \
  --subject "Your August account update" \
  --preview-text "A quick look at what changed this month." \
  --sending-purpose marketing

Use --file - to read HTML from stdin. Use --input - to read the complete JSON request from stdin.

Run an Audit from the TypeScript SDK

const result = await brew.emails.auditEmail({
  emailHtml,
  subject: 'Your August account update',
  previewText: 'A quick look at what changed this month.',
  sendingPurpose: 'marketing',
})

if (result.completion.status === 'partial') {
  console.log('Audit incomplete. Retry before claiming readiness.')
} else {
  console.log(result.completion.readiness, result.completion.score)
}

Pass { raw: true } as the second argument to read X-Credit-Cost and the other response headers.

Run an Audit from the API

POST /v1/emails/audit accepts raw content rather than a saved email id:

curl "https://brew.new/api/v1/emails/audit" \
  -H "Authorization: Bearer $BREW_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: audit-august-newsletter" \
  -d '{
    "emailHtml": "<!doctype html><html lang=\"en\"><body>...</body></html>",
    "subject": "Your August account update",
    "previewText": "A quick look at what changed this month.",
    "sendingPurpose": "marketing"
  }'

The endpoint returns 200 for both complete and partial audits. Reuse the same idempotency key when retrying a partial result. See the generated API reference for the complete versioned schema and transport limits. Audit calls have a separate admission policy. See Rate Limits for the current caller, organization, and concurrency limits.

Checks That Need Send Context

An HTML audit cannot verify SPF, DKIM, DMARC, one-click unsubscribe headers, the plaintext MIME part, consent or suppression state, or inbox placement. Use domain health, contact validation, and an inbox placement test for those checks. Before every send, transactional creation, and export, Brew runs the free content checks again on the exact email. Missing online results alone do not stop the action when an audit is missing, stale, or partial. Deterministic blockers still stop it. Marketing sends keep Brew’s unsubscribe insertion as the last safety check.

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:

Self-Service Tools
Talk to Our Team

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.

Get started

Emails

Before You Send

Brand

Audience

Analytics

Settings

Security and Compliance

Run an Audit in the App

Understand the Result

Choose the Sending Purpose

Continue After a Warning

Run an Audit from MCP

Run an Audit from the CLI

Run an Audit from the TypeScript SDK

Run an Audit from the API

Checks That Need Send Context

Need Help?

Search Documentation

ChatGPT/Claude Integration

Schedule a Call

Call Us Directly

Slack Channel

Email Support

​Run an Audit in the App

​Understand the Result

​Choose the Sending Purpose

​Continue After a Warning

​Run an Audit from MCP

​Run an Audit from the CLI

​Run an Audit from the TypeScript SDK

​Run an Audit from the API

​Checks That Need Send Context

​Need Help?

Search Documentation

ChatGPT/Claude Integration

Schedule a Call

Call Us Directly

Slack Channel

Email Support

Run an Audit in the App

Understand the Result

Choose the Sending Purpose

Continue After a Warning

Run an Audit from MCP

Run an Audit from the CLI

Run an Audit from the TypeScript SDK

Run an Audit from the API

Checks That Need Send Context

Need Help?