← all playbooks

Check your sending mailboxes and domain before a campaign goes out

A go or no-go on every mailbox you plan to send from, each with its reason: mailbox state, warm-up, authentication records and bounce history. Plus what the healthy mailboxes can carry between them, so the send is sized to them. Nothing in your sequencer is changed.

use when
a campaign is loaded and ready, and you want to know the mailboxes and the domain behind it are fit to send before you press start
starts from
A list

The prompt

Paste it into Claude Code or the Claude desktop app with Hyreflow connected. The first line tells your agent to use Hyreflow, so it reads the play, asks before it spends anything, and hands the work back to you.

paste this into Claude
Use hyreflow and check that my sending mailboxes and my domain are healthy
before I start this campaign.

The campaign is <CAMPAIGN_NAME> in <SEQUENCER>. It goes to about
<LIST_SIZE> people from these mailboxes: <MAILBOXES_OR_ALL_ON_THE_CAMPAIGN>.

For each mailbox give me go or no-go, with the reason:
- Is it active, is sending switched on, is it in an error state?
- Warm-up: not started, in progress, paused or complete, and where the
  warm-up mail has been landing.
- Authentication: the MX, SPF, DKIM and DMARC records, passing or not.
- Bounce history on my last <NUMBER_OF_CAMPAIGNS> campaigns.
My rule for a no-go is <MY_RULE_OR_USE_YOURS>.

Then tell me how many sends a day the go mailboxes can carry between
them, and how that compares with this list.

Check one mailbox first and show me the verdict, and tell me what the full
run costs before you do the rest. Read and report only. Do not start,
pause or change anything, and ask me before you run any test that sends
seed mail from my mailboxes.

Replace every <PLACEHOLDER> with your own detail. Everything else can stay as written.

What you need first

  • The campaign you are about to start, or the list and the mailboxes you plan to send it from
  • Your sequencer connected with your own account. SendKit gives the full check, the others give part of it
  • Your own rule for a no-go, or agreement to use the play's defaults

Tools it can reach for

The agent picks per step from what your workspace has. Nothing here is required by name.

What happens when you run it

Free steps are marked free. Anything that spends credits is marked, and the agent asks before the first paid run of any size.

  1. 1

    Find the mailboxes behind the campaign

    free

    In SendKit the mailboxes assigned to the campaign are listed directly. Smartlead and Reply.io list the sending mailboxes on the account, and Reply.io also shows which ones are linked to a sequence. Anywhere else, you name them.

  2. 2

    Read each mailbox's state

    free

    Active, inactive, suspended, failed or in an error state, and whether sending is switched on. A mailbox that is not active and enabled is a no-go before anything else is looked at.

  3. 3

    Read the warm-up

    free

    Whether warm-up was never started, is in progress, paused, stopped or complete, and how the warm-up mail has been placed: inbox, spam or promotions. A mailbox that finished warming last week is a different risk from one that has been sending for months.

  4. 4

    Check the authentication records

    free

    The MX, SPF, DKIM and DMARC status is read for every mailbox and grouped by domain, because a broken record is a domain problem: every mailbox on that domain shares it.

  5. 5

    Read the bounce history

    free

    Bounce counts on your recent campaigns are read from the sequencer, so a mailbox or a domain that has already been bouncing is visible before it carries another send.

  6. 6

    Run a placement or blacklist test, only if you approve

    free

    Past test results are read for free. A fresh inbox placement test sends seed mail from your mailboxes, and both tests count against your SendKit plan's monthly allowance, so the agent asks before it creates either.

  7. 7

    Give the verdict and size the send

    free

    Every mailbox gets go or no-go with its reason in plain words. The go mailboxes' daily capacity is added up and set against the list, so you know how long the send takes on healthy mailboxes only.

What a run costs

Nothing in this play spends credits.

Free before anything is charged

  • Find the mailboxes behind the campaign
  • Read each mailbox's state
  • Read the warm-up
  • Check the authentication records
  • Read the bounce history
  • Run a placement or blacklist test, only if you approve
  • Give the verdict and size the send

What moves the number

  • How many records survive the free filters. Everything dropped before the paid steps costs nothing.
  • Providers you connect with your own key. Those calls bill your account, not your credits.

An estimate, not a quote, priced at the volume credit rate. Your agent sizes the run against your own workspace and tells you what it will cost before it spends anything.

A clean list sent from a cold mailbox still lands in spam

Deliverability has two halves. One is who you send to, and checking the list before you send covers it. The other is what you send from, and it gets far less attention, because nothing about it is visible from inside a campaign.

Mailbox providers judge the sender before they read the message. Is the domain authenticated? Has this mailbox built up a history of ordinary mail, or did it appear last week and start writing to strangers? Is it throwing errors? A campaign started on a mailbox that only just finished warming, or on a domain whose DKIM record broke when somebody moved the DNS, does not fail loudly. Nothing bounces. There is no error. The mail goes to spam, the replies do not come, and you blame the copy.

This play is the check you run while the campaign is still idle. It reads the state of every mailbox you plan to use and tells you which ones should carry the send and which should sit it out.

What you get back

One row per mailbox: go or no-go, the reason in plain words, its state, its warm-up status and where warm-up mail has been landing, and the four authentication records. One row per domain, because a failing record takes every mailbox on that domain with it. The bounce picture from your recent campaigns. And a capacity line: what the go mailboxes can send in a day between them, against the size of the list.

Nothing is started, paused or edited.

Variations worth knowing

Run the placement test. On SendKit the agent can create an inbox placement test, which sends seed mail from your mailboxes and reports where it landed. It asks first, and it reads past results for free.

Not on SendKit. Paste what your tool shows for warm-up and domain records, and the same rules produce the same verdicts.

Then build. With the go mailboxes named, build the outreach sequence and attach only those.

Where this goes wrong

A go is a snapshot. It describes the mailbox on the day it was read. If the campaign slips by a month, check again.

Warm-up complete is not a licence for full volume. A mailbox that has only ever sent warm-up mail should ramp up, not open at its daily maximum.

One domain, one reputation. Ten healthy mailboxes on a domain with a failing DMARC record are ten no-gos. Fix the record first.

Sending from the domain your clients write to. If campaign mailboxes share a domain with your everyday email, damage to one reaches the other.

Passing every check with a bad message. Healthy mailboxes do not protect you from complaints. The check says nothing about whether the reader wanted the email.

Questions

Which sequencers can it check?

SendKit fully: mailbox state, warm-up progress, where warm-up mail lands, the MX, SPF, DKIM and DMARC records, and inbox placement and blacklist tests. In Smartlead and Reply.io the connection lists your sending mailboxes but does not read warm-up or DNS. In Instantly and lemlist it reads campaigns and their bounce figures, not the mailboxes. For those, read the warm-up and domain screens in the tool itself and paste what you see: the agent applies the same go or no-go rules to it.

What makes a mailbox a no-go?

Any one of these: the mailbox is inactive, suspended, failed or in an error state; sending is switched off; warm-up was never started, is paused or stopped, or its health is reported as poor; one of the four DNS records is failing; or the sequencer itself flags the bounce rate as high. Those are the defaults. If your own rule is stricter or looser, say so in the prompt and that rule is applied.

Does it fix what it finds?

No. It reads and reports. DNS records are corrected where your domain is hosted, not in the sequencer. In SendKit the agent can start or resume warm-up on a mailbox, but that puts warm-up mail on the wire, so it happens only when you ask for it and approve it, never as part of the check.

Does it cost credits?

No Hyreflow credits. Every sequencer runs on your own account with them, and reading your own mailboxes and campaigns is free. The two SendKit tests are limited by your SendKit plan, which is why they wait for your approval.

Can it run every Monday?

Yes, once you have run it by hand and agree with the verdicts. It converts into a scheduled workflow that reads the mailboxes and leaves the go or no-go report waiting in your workspace. A scheduled run reads only: it creates no tests, changes nothing and sends nothing. Prove it by hand first, so the no-go rule is one you trust before it runs unattended.