Choose what you want to protect
Photos, documents, archives, backups, and other important files.
7IRIS Vault · private testing in preparation
7IRIS Vault is being built for photos, documents, and archives that cannot be replaced. The laboratory core encrypts a file before splitting and reconstructs it from 6 admissible parts out of 10. The user application is still being prepared for private testing.
Example recovery check: 03:14
All parts are available
In the 7IRIS model, the decryption key remains with the owner; this user flow still needs application-level verification.
This is an example of the planned app interface.Planned recovery after losing a computer
This is the planned flow for the first 7IRIS application; it has not been released to users yet.
Test it yourself
Switch off several nodes and see that the data can still be recovered. This demonstrates the “6 of 10” principle; these are not live network nodes.
10 of 10 · demo complete
Three steps in the planned protection flow
Photos, documents, archives, backups, and other important files.
The laboratory core encrypts a file before splitting it into parts. The future app must keep plaintext and the key on the owner’s device only.
In the laboratory, the core reconstructs data and compares the result. Scheduled checks in the user application are still planned.
One direction. Three planned roles.
The first product for protecting irreplaceable data: photos, documents, backups, and other important files.
The plan is for family and friends’ computers to help store encrypted parts without access to their contents.
A future Node will store encrypted parts only. Its owner will choose storage, speed, and operating hours.
Not just another cloud
7IRIS Vault is being designed for long-term protection of data that is difficult or impossible to recreate. Instead of depending on one disk or cloud provider, encrypted parts should be stored across several locations. The laboratory profile recovers from any 6 correct parts out of 10; the distributed network and standalone recovery are still in development.
Memories that cannot be photographed again.
Copies of documents and important records.
Backup keys, recovery codes, and other critical data.
A project archive and years of work.
Contracts, accounts, and critical backups.
A future scenario for giving loved ones access to important data under rules set in advance. The feature is still planned.
Long-term direction
In the laboratory model, plaintext and the key are not sent to storage nodes. The user application still needs to verify this boundary. Vault remains the first product; transfer, digital legacy, and other scenarios are future directions.
Photos, documents, archives, and backups. A Windows app, recovery key, and independent offline recovery tool are the MVP priority.
A planned one-time encrypted file transfer without a permanent link. Private routes and layered encryption are still being researched.
A future scenario for protecting encrypted data needed to recover access to digital assets, accounts, and other important services. The designed feature should not hold money, manage cryptoassets, or gain access to the owner’s keys.
Planned digital inheritance: an encrypted capsule containing documents, family archives, and instructions is intended to be released to loved ones only after a waiting period and multiple confirmations. 7IRIS does not transfer assets or replace a will.
A separate future vault for logins, passwords, recovery codes, and access keys.
A one-time code will allow the encrypted file to be retrieved once and then stop working. After saving the file, the recipient controls their copy — 7IRIS cannot delete it from their device. Private routes remain under research; absolute anonymity is not promised.
Dedicated direction page →A future protocol will collect independent evidence of human presence and authorization for a specific action without creating a biometric database of its own.
How strongly external credentials support that this is one unique person.
How strongly the evidence supports that a person was present for the action.
How clearly human consent is bound to this exact action.
Declared mode: Human, Human + AI, or an autonomous agent authorized by a human.
Which evidence was used: a signature, hardware key, challenge-response, or other signals.
The proposed protocol should not claim to determine whether AI helped a person. It is intended to provide a verifiable confidence level for presence and authorization. Personhood should not be a mandatory condition for an owner to access data in 7IRIS Vault.
Some technical data, such as connection time, encrypted file size, or an IP address, may remain visible at the network level. We do not promise absolute anonymity; our goal is to get as close to it as modern technology allows.
Our principles
The laboratory core does not send plaintext or the key to storage locations; the user application still needs to verify this boundary.
Private Transfer is researched separately and does not promise absolute anonymity.
The designed model does not give the decryption key to the 7IRIS team or storage nodes.
The laboratory core already reconstructs a test copy and compares the result; regular application scheduling is still being designed.
Honest status on August 3, 2026
The app is not available yet. We are recruiting private-test participants, not presenting the prototype as a finished service. Technical details →
Private testing
We are looking for early users and engineers willing to test the first Windows build once it is ready and help make recovery more reliable.
Short answers
The architectural goal is company-independent recovery through a documented format, encrypted-part export, and a standalone utility. These capabilities are still being developed and require separate verification; they are not presented as a finished guarantee.
With ordinary cloud storage, access depends on one provider's account and infrastructure. 7IRIS is being designed differently: a file is encrypted on the device, parts are distributed across several locations, and the laboratory core already verifies 6-of-10 reconstruction with SHA-256. The distributed network, part export, and standalone recovery are still in development.
No technology or storage system can guarantee absolute invulnerability. 7IRIS is designed to reduce dependence on any one device, storage location, or company. Key custody, device security, the availability of enough parts, and other external risks still matter.
In the verified laboratory profile, a file can be reconstructed exactly from any 6 valid parts out of 10. This means that up to 4 unavailable parts did not prevent recovery in the tests. It is a laboratory result, not a guarantee against every possible combination of failures.
During laboratory tests, the core reconstructs a test file and compares its SHA-256 with the original value. This verifies reconstruction accuracy in that test. Continuous automatic monitoring of real user archives has not been released yet.
The laboratory core encrypts a file before splitting it into parts. Plaintext and the key are not sent to nodes. The final key model and cryptographic format will be published after the client design and verification are complete.
Primarily for the long-term protection of hard-to-replace data: family photos and videos, documents, project archives, recovery codes, and other digital valuables. Vault is not presented as a ready replacement for everyday work-file synchronisation.
The planned flow is to install 7IRIS on a new device, confirm access with a recovery key, obtain enough encrypted parts, and verify the reconstructed result. This user flow has not yet been released.
Home nodes are planned to be supplemented by stable reserve nodes. Recovery requires 6 valid parts out of 10. Creating new parts after an outage is a future app feature; its automation is still being developed.
Node is designed to use one dedicated folder under strict resource limits. It should store opaque encrypted parts and never execute them. The physical client still needs to verify this boundary.
Not yet. We are preparing the first Windows build and recruiting private-test participants through the website form.
7IRIS must not have a hidden way to decrypt data without the owner’s key. During setup, the future client will require users to create and test a separate recovery key.
Submitting an application creates no financial obligation. There are no tokens, cryptocurrency, or income promises. Testing terms will be provided before participation is confirmed.
Contact
Questions about private testing, architecture, and recovery checks — on Telegram.
Technical details →