All packsTransactional Email Deliverability CoreSolve
email deferred send never retried dropped silently
A deferred send with nothing replaying it is a dropped send
Warmup correctly defers a send that would exceed the daily cap -- but a deferral with no drain routine to replay it later is functionally a silent drop.
This is one of the things Transactional Email Deliverability Core already handles. Send transactional email that actually lands, and stop emailing people who bounce or complain.
Is this you?
Deferring a send that would exceed the day's warmup cap is the right call -- but deferring only helps if something eventually retries it. A system that records a deferred row and stops there has, in practice, dropped that email: the password-reset or receipt the user was waiting on never arrives, and nothing in the deferral itself schedules a follow-up attempt.
Why this one is easy to get wrong
Implementing the deferral check ("if over cap, don't send, record it") reads as a complete feature on its own, and it passes any test that only checks "a send over the cap doesn't go out." Building the replay routine is a second, separate piece of work that isn't implied by the deferral logic itself -- it's easy to ship the half that prevents over-sending and never build the half that un-defers anything.
What you get instead
The pack ships drainDeferred() as a first-class routine, not an afterthought: it replays deferred rows in FIFO order (oldest first, so a busy day's early senders don't starve behind later ones), stays budget-bounded against the same day's remaining warmup allowance, and marks each replayed row with replayed_at so a second drain run -- cron overlap, manual retry -- can't double-send. GO-LIVE-CHECKLIST.md has the buyer schedule it on day one, specifically so nothing sits deferred waiting on a script that was never scheduled.
Source: ARCHITECTURE.md 'Deferred-send drain' — checkable in the pack you receive
How you actually use this
You don’t install a library or wire up an SDK. Your own coding agent builds the code in your project, and you keep it — no runtime dependency on us.
Step 1
Download and unzip
You get a folder: the docs that tell an agent what to build, a starting skeleton, and the test suite that decides when it's done.
Step 2
Open it in Claude Code or Cursor
Point your coding agent at the folder. Nothing to install, no account with us, no API key.
Step 3
Paste one prompt
The pack contains the exact prompt. Paste it as your first message and leave it alone — it works through the build itself, choosing a cheaper or stronger model per task.
Step 4
Run ./verify.sh
One command. It prints a pass or fail for every check. Green means the build is done — the same script we ran to produce the receipt on this page.
Typical build: about 13 minutes of your agent working, mostly unattended. Then you integrate the working module into your app the way you would any code you’d written yourself.
Why you can believe this
3 of 3 runs passedWe ran this pack from an empty folder 3 times and published exactly what happened — every check, the model, the token cost, the wall time. Not a testimonial, and not our opinion: the same verify.sh you run yourself. Read the full receipt →