When both correspondents use Crypted.li, the protected-mail architecture is designed so Crypted.li itself does not retain the key required to decrypt message content. The objective is simple: a request made to our servers — including by a state authority — cannot produce plaintext that the service has no cryptographic ability to recover.
BUILD THE VAULT ↓The strongest Crypted model begins when both sender and recipient use Crypted.li. The conversation can remain inside a system designed around protected content rather than routine inspection.
Crypted.li is designed in opposition to routine mass scanning of private correspondence. Protected Crypted-to-Crypted mail is not intended to become a general-purpose surveillance feed.
The protected payload is sealed before server storage and transport. The infrastructure handles encrypted data rather than relying on permanent access to readable message content.
Message-specific cryptographic material exists only for the operation that requires it. The design must avoid a permanent server master key capable of reopening every protected conversation.
The target architecture keeps the long-term decryption capability outside the mail server. Crypted.li should transport and store protected content without retaining what is needed to turn it back into plaintext.
With both endpoints on Crypted.li, no external mailbox provider needs to receive the protected conversation. This is the route designed for the highest level of control.
The recipient side receives the cryptographic capability required by the final client protocol. The mail server itself is not supposed to become a second permanent reader of the conversation.
Temporary service-side decoding material is destroyed as soon as its role is complete. Backups, indexes, logs and caches must preserve the same zero-access property.
If Crypted.li does not possess a usable decryption key, a demand directed at Crypted.li cannot make the service produce plaintext it cannot cryptographically recover. Endpoint compromise remains a separate threat.
When both endpoints use Crypted.li, the final protocol is intended to keep protected message content unreadable to the mail service itself. No external mailbox provider needs to enter the delivery path.
Standard email remains possible, but the zero-access Crypted-to-Crypted guarantee cannot automatically extend into another provider. Their infrastructure, policies and legal environment become part of the trust chain.
Crypted.li is intentionally narrow. It is being designed as private mail, not as another cloud platform that happens to contain an inbox.
Protected Crypted-to-Crypted correspondence is designed not to be available to Crypted.li as plaintext for routine scanning, advertising profiling or indiscriminate server-side inspection.
The service deliberately refuses unnecessary media storage. Less accumulated content means less data available to expose.
150, 250 or 350 MB. The product encourages communication rather than endless accumulation.
The production client protocol, key exchange, storage, backups, indexing, notifications, logs and caches must all preserve the same rule: Crypted.li must not retain a usable path back to plaintext.
Every tier gets the same Crypted experience. Upgrades remain flexible; downgrades become available once usage fits inside the target quota.
150 MB mailbox
No image/video attachments
Web + native apps
250 MB mailbox
No image/video attachments
Web + native apps
350 MB mailbox
No image/video attachments
Web + native apps
Keep both correspondents on Crypted.li. The protected-mail protocol is designed so Crypted.li does not retain the key required to decrypt the conversation. That means the service itself is not intended to be capable of producing plaintext from its servers later — including in response to a state demand directed at Crypted.li. Security of the users' own devices remains a separate endpoint responsibility.
CREATE PRIVATE MAILBOX ↗