How Temp Mail Works: From MX Records to Your Inbox
TempMailPortal creates a short-lived private mailbox record when you open a new random inbox. The receiving domain routes mail to a Worker, which accepts a message only when its recipient is eligible. Follow one message from the sender to the screen and the setup becomes much less mysterious.
Step 1: You ask the browser for a fresh address
On a first visit, the site waits until you select Create inbox. The Worker then creates a random local part—the characters before the @—and pairs it with an active disposable-mail domain. An illustrative address format is noah.baker123456@namesgeneratorhub.com; do not use this example as a real inbox. There is no permanent user password or recovery profile behind a guest web inbox. The response contains the address and a private mailbox credential, which the browser keeps locally for later reads. If that browser already holds an active inbox credential, reopening the page restores the same address rather than allocating another one.
The browser now creates private random inboxes only; it no longer opens legacy public names. Address access lasts separately from message retention. Keep the browser profile and pairing codes secure, and use temporary mail only for permitted messages you can afford to lose.
Step 2: The sender finds the receiving servers
The website sending the email treats the temporary address like any other destination. Its mail server looks up the domain's MX records in DNS to find out where incoming mail should go. Think of an MX record as the delivery sign for email: it tells the sender which receiving service handles mail for that domain. Here, the records point to Cloudflare Email Routing for the disposable domain. The sender does not need to know that the destination is temporary; it only needs a valid domain whose receiving servers accept the message.
This is why the brand domain and mailbox domains are kept separate. tempmailportal.com hosts the website and API, while disposable mail is received on domains intended for that purpose.
Step 3: A catch-all passes the message to the Worker
A conventional mail server often rejects an address it has never seen. A disposable domain instead uses a catch-all route to send mail for otherwise unnamed local parts to application code. The Worker then checks the recipient against its current mailbox rules; private-looking addresses with no active mailbox record are rejected. Cloudflare documents catch-all routing in its Email Routing address reference.
The Worker checks that the recipient belongs to a configured mailbox domain, parses the raw MIME message and separates the sender, subject, text, HTML and attachment details. Unsafe HTML features are removed before the body is returned to the browser. The remaining message data is stored against the exact mailbox local part with an expiration time.
The service does not pre-register thousands of addresses in DNS. One catch-all route passes incoming mail to the Worker, which checks for an active private mailbox before storing a message for it.
Step 4: The open page checks for new mail
The inbox page sends its private credential to the service API on a short interval and asks for message envelopes. Once a message has been stored, its sender, subject and preview appear without a manual refresh. Opening that row triggers a second request for the full body. In ordinary conditions the visible delay is the sender's delivery time plus whatever remains before the next poll; it is not a guaranteed instant-delivery service.
The credential limits reads to one exact private mailbox. Knowing the address alone does not open it, but anyone with access to the browser profile or a valid pairing code can gain access. Lost credentials cannot be reset. That is why the interface is suitable for low-stakes mail and manual tests, not secrets or account recovery.
Step 5: Scheduled cleanup removes old messages
Every stored message carries an expiration timestamp. A scheduled Worker task deletes expired database rows and any retained attachment objects. The selected retention is 10 minutes, 1 hour or 24 hours after each message arrives, capped by the mailbox's own expiry. There is no archive or restore button after cleanup, and closing an inbox can remove its messages sooner.
The address itself is not a permanent asset either. A browser may remember the current address and token for convenience, but clearing that local state or switching devices can leave you without the token you were using. If a message or account will matter tomorrow, a temporary inbox is the wrong place for it.
Where the address, message and file actually live
A temporary email is not one object that disappears all at once. The current browser and Worker implementation keep several pieces separately. This map explains why changing an address, forgetting it locally and deleting the server inbox produce different results.
| Piece | Where TempMailPortal uses it | What ending access means |
|---|---|---|
| Visible address and private browser credential | The current browser profile stores them so it can reopen that inbox without a Google sign-in. | Clearing browser data loses your access; it does not itself send a server deletion request. |
| Mailbox record and selected retention | The Worker records the exact address, its expiry and whether each message should live 10 minutes, 1 hour or 24 hours. | The browser mailbox can remain eligible for new mail for up to 7 days even after earlier messages expire. |
| Message text and metadata | The receiving Worker stores them against the exact mailbox in its database, with an expiry timestamp. | The API stops returning the message at expiry; scheduled cleanup reclaims the stored row afterward. |
| Retained attachment object | The file is held in private object storage while its message is active. The website can display attachment details but cannot download the file. | Deleting the inbox requests removal of retained objects; the verified Android app is the download path while access remains valid. |
Checked against the deployed Worker on 3 November 2026: the website's first-visit path waits for an explicit create action, and a local browser test confirmed that Delete returns to an empty start state without creating a replacement. The production attachment route returned HTTP 403 to an unauthenticated request without browser headers. These checks do not measure external sender delivery, a 24-hour expiry boundary, or a download on a Play-installed phone. For the limits of private access, read the temporary-email safety guide.
What we verified in the Worker
Historical check, 28 August 2026: we passed a crafted MIME message through the same email() handler used by Email Routing. The expiry observation below used the then-default 24-hour retention; it does not describe every retention option now available. The fixture deliberately contained useful text alongside a script, an iframe, a form and password-style input, an inline event handler, a javascript: link and an ordinary HTTPS link.
| Check | Observed result |
|---|---|
| Mailbox-domain boundary | namesgeneratorhub.com was accepted; the brand domain was rejected |
| Stored expiry in that fixture | The stored expiry was exactly 24 hours after receipt under the default used in that test |
| Active and embedded content | The script and iframe were removed |
| Credential-capture markup | Form tags and the input element were removed |
| Attribute-based execution | The inline event handler was removed and the javascript: URL became # |
| Legitimate content | The message text and ordinary HTTPS link remained |
This historical fixture predates the private-inbox update and used the then-default 24-hour lifetime. It is an application-layer check, not a claim that an email is harmless after sanitisation. A safe-looking link can still lead to a phishing site, and this fixture did not measure delivery from an outside mail provider through DNS and MX routing.
Questions people reasonably ask
How quickly should a message appear? There is no fixed delivery time: it depends on the sender, DNS, the receiving path and the next browser poll. The implementation check above tested parsing and storage, not external delivery latency. If nothing arrives, first confirm that the sender used the exact address and that its own system actually sent the message.
Can somebody else read the inbox? The address alone is not enough to open a new private inbox. Someone with the browser credential, access to your browser profile, or a pairing code can read it. Never send personal documents, recovery links or important credentials to a throwaway inbox.
Can TempMailPortal send or reply? No. The service is receive-only. A message may show a reply-to address, but there is no composer or outbound delivery path in the product.
Is this the same as an email alias? No. An alias normally forwards to a permanent mailbox and stays useful for future mail. A temporary inbox stores mail separately and is designed to be abandoned. The broader choice is covered in disposable email vs. aliases.