Sign in to
your relay address portal
Enter the email address you used to claim your relay address. We'll send you a sign-in link - no password needed.
We'll send a one-time sign-in link to this address. The link expires after 5 minutes.
Recover
your account
Enter your recovery code and the new email address you want to use from now on. We'll move your account and send a sign-in link to the new address.
Your recovery code is single-use. After recovery you'll be asked to save a new one.
Check your inbox
If that address has a RELAY account, we've sent it a sign-in link and a 6-digit code.
Or just click the link in the email. Both the link and the code expire after 5 minutes.
Save your recovery code
This is the only way back into your account if you ever lose access to your sign-in email. RELAY stores it in a form we cannot read, so we can never recover it for you. Save it now - you will not see it again.
📋 Account details
🔗 Sub-addresses
Context-specific identities visibly linked to your canonical address, written as yourname.relay@suffix. Anyone who sees one knows it belongs to you. Your canonical always stays the root.
⚙️ Routing preferences
Senders only ever know your relay address. Map each app to one of your contact points (managed under Account details) and set the order RELAY tries them in.
Your routing rules are private - never shared with senders. A channel goes live once its contact point is verified.
🛡️ Who can reach you
Rules for your main relay address. Senders never see these - the routing engine applies them when it delivers. Enforcement goes live with the routing engine.
These rules are private and never shared with senders. They take effect once the routing engine is live.
✉️ Contact requests
People reaching you for the first time wait here until you let them in. Approve to receive everything they sent and allow them from now on; reject to discard their messages and block them. Both can be undone below.
Your relay name, written in the alphabet you use every day. RELAY will render your canonical name as a transliteration in the script you choose - the same name and the same identity, simply written in your own writing system. Always generated from your canonical address, never a different name.
When you create an AI agent, a unique key pair is generated: a private key (which you downloaded and keep safe) and a public key (which RELAY stores to identify your agent).
The key fingerprint is a short code derived from the public key - like a serial number on a passport. It lets you quickly check that a key file on your device really belongs to this specific agent.
In practice: if you ever have multiple agents and are not sure which key file belongs to which agent, you can compare the fingerprint shown here with the fingerprint of the file on your device. They should match.
For most people, most of the time, you will never need to check this. It is there as a safety net.
Your canonical address is your permanent, unique relay address - the root of everything. Think of it like a home address that never changes: you can redecorate, move furniture, or get a new phone number, but the address on your front door stays the same.
Anyone who knows your canonical address can always reach you, even when the channels behind it change. You decide which app or inbox each message arrives in - the sender never needs to know.
Every sub-address you create (@work, @agent, @family) is linked to this canonical. It is the one identity that is always, permanently yours.
Your canonical is your one permanent public door: you give it out widely and can swap the channels behind it without anyone noticing. A sub-address adds what the canonical cannot - a separate door with its own rules and its own on/off switch, and, for agents, provable origin.
Private (@family, @friends, your choice)
You give your canonical out widely, which might build up some noise - and you will then put rules on it: strangers must ask permission first, and their messages wait in a separate queue. But your partner, your child and your doctor get yourname.relay@family. That address is set up to come straight through to your phone with priority, past every filter, even at night. You stay unreachable to the noise and instantly reachable to the people who are closest to you. With your canonical alone, everyone comes through the same door under the same rules - you cannot make that distinction.
Work (@employer, issued by your employer)
While you are there, the address carries the backing of a verified organisation - the recipient sees you really are writing on that company's behalf. You have spent years building client relationships, and everyone knows you through work. On your last day, yourname.relay@employer is cleanly retired - contacts who only had your work address know it is no longer active. But your canonical stays yours: anyone who needs you personally still finds you. Your network travels with you instead of with your employer.
Company (@yourcompany, verified company)
This is not about being able to switch channels - your canonical already does that. It is about verified trust. A client who receives an invoice from yourname.relay@yourcompany knows, cryptographically, that this is the registered company behind you - not a fraudster imitating your firm to send a fake invoice. Your canonical says "you"; your company address says "you, provably acting on behalf of this verified company". In an age of invoice fraud and spoofing, that is not tidiness but protection - for you and for your client.
Press (@press, issued by an organisation)
A media organisation - a newspaper, broadcaster or publisher - can issue @press to its journalists and editors. Only verified media organisations can do this. If you have been issued a @press address by your employer, you can use it as proof of credentials when reaching out to sources, PRs or public figures. And on the receiving end, anyone who gets a message from a @press relay address knows it is a verified journalist from a recognised organisation - and can set their own routing preferences accordingly: press enquiries straight to a dedicated channel, automatically prioritised, or routed directly to WhatsApp. The credibility is built into the address itself.
AI agent (@agent, @scheduler)
Agents will act on your behalf: making appointments, ordering, negotiating, making commitments. The recipient needs to be sure of three things - did you authorise this, is this an agent or you yourself, and does that check out cryptographically? Today an AI using your email address is indistinguishable from you; if that agent slips up or is hijacked, you carry the blame with no clean way to deny it. With yourname.relay@agent it is a separate, key-bearing identity: you revoke that one agent with a single click (a kill switch), the recipient verifies it is an agent you sanctioned, and "a human did this" versus "your agent did this" becomes provable. Your canonical is you; your agent address is provably-from-you-but-not-you. As agents start handling money and commitments, that distinction becomes a necessity, not a luxury.
This is the only time your agent's private key is shown. RELAY never stores it - it lives only on your device. Anyone who holds it could potentially act as this agent, so treat it like a password: store it somewhere safe. If you lose it, the agent address can no longer be used and you will need to create a new one.
If you prefer a recovery path, RELAY can store your key as an encrypted backup. You choose a passphrase; the key is encrypted in your browser before it is sent. RELAY only ever holds the encrypted blob and cannot read it without your passphrase. If you forget the passphrase, the backup cannot be recovered.
Enter the passphrase you chose when you created this backup.