How our encryption works
OnePeek is built so that we can’t read what you share. Everything is encrypted in your browser before it’s uploaded, and the key never reaches our servers.
Last updated 26 October 2026
Real encryption, right here
This is the same process that runs when you send a secret. Type anything and watch your browser generate a key, encrypt it and build the link. Then it decrypts it again to prove the key is all you need.
- Runs entirely on your device
- A new IV for every encryption
- 16-byte tag catches any tampering
Nothing you type here is sent anywhere. This uses the same Web Crypto API OnePeek uses.
………5 · The link you’d shareEncrypting…
Four steps, all in your browser
- 01A fresh key for every secret
Your browser creates a random 256-bit key with the Web Crypto API. Keys are never reused.
- 02Encrypted on your device
Text, files and Wi-Fi details are encrypted with AES-256-GCM and a new random 96-bit IV. The file name, type, size and viewing timer are encrypted too.
- 03Only ciphertext is uploaded
We store scrambled bytes. GCM also detects tampering, so altered ciphertext won’t decrypt.
- 04The key travels only in the link
It sits after the
#. Browsers never send that part to a server, so it isn’t in our logs either.
Every lock is checked before a single byte leaves
The viewer’s browser asks our server for the ciphertext. First, the server checks every lock the sender chose. On the last allowed view the ciphertext is deleted from the database and encrypted files are deleted from storage.
The browser then decrypts locally with the key from the link and removes the key from the address bar.
Encrypted to a key only you hold
When you ask someone for a password, your browser makes an ECDH P-256 key pair. The public half goes into the request; the private half stays in your browser (and in an optional private link only you keep). The answer is deleted as soon as you read it.
What we can and can’t see
What we store
- Ciphertext, which we can’t decrypt
- When a secret was created, when it expires, and how many views it allows and has used
- Settings needed to enforce your locks: recipient email, opening times, allowed countries and the geofence area
- If you’re signed in: your history, including who you noted it was for, whether it was opened and, if you asked, roughly where
What we never see
- Your secret, file or Wi-Fi password
- The key, or the part of the link after
# - File names and types inside the secret
- Answers to your requests, which only your private key can open
Limits you should know about
Anyone with the full link can open the secret. Recipient lock, time lock, country and geofence add checks on top.
For sensitive secrets, use a recipient lock or send the link over a different channel.
“Blur until held” and the watermark discourage screenshots, but nothing stops someone photographing a screen.
Geofences trust the location the device reports. Country checks use the network address, which a VPN can change.
If malware runs on the sender’s or viewer’s device, it can read what’s on screen.
See it in your own browser
Open your browser’s developer tools, go to the Network tab and send or open a secret. You’ll only see ciphertext going up or coming down, and the part of the link after # never appears in any request.
Report a security issue
If you find a vulnerability, please email secret@onepeek.io with steps to reproduce it. Please give us a reasonable time to fix it before sharing it publicly, and don’t access other people’s data while testing. Our security.txt has the same details.