Security & privacy
Little to steal. Nothing to hide.
Reachy connects you with strangers on the street and knows where you park. So we built it to collect as little as possible and to lock down the rest. This page explains what that means for you, in plain words.
They never see who you are.
No name, no phone number, no plate. They can only send "please move your car" — and you answer with a button.
We don't keep a history of your trips.
Your phone uses your location to help you park. We don't build a map of where you've been.
We never sell it. To anyone.
Not to advertisers, not to tow companies, not to data brokers. We make money from Premium and the Reachy Store.
our promises
What Reachy never keeps
These aren't settings you have to find. Reachy simply has no place to store them.
Free spots you leave
When you pull out, the spot shows on the map without your name or account — and it's deleted after 15 minutes.
Map problems you report
The spot and a picture of the map, kept for 30 days so we can tell you when it's fixed.
Your permission, on record
We only message you because you agreed to it — and we keep proof of that. Turn your badge off or delete your account any time.
how we protect it
Three locks on everything that's yours.
We assume someone will one day try to break in. So even if they get past one lock, two more are waiting.
Encrypted, like online banking
Everything between your phone and Reachy travels encrypted. Nobody on the same Wi-Fi can read it.
Codes that expire fast
Sign-in codes last 5 minutes, and too many wrong tries are blocked. Sign in with Apple or Google works too.
Locked in a vault
The passwords that run Reachy are kept in an encrypted vault only the founders can open, and changed regularly.
Alerts that give nothing away
"You got a new violation." "Someone needs to get out." Never your plate, the amount or a stranger's details.
A random code, not your ID
The code on your badge means nothing on its own. Lose it? Switch it off in one tap and get a new one.
Help without snooping
Our team finds problems using a reference number, not by looking through your personal information.
your car
Plugged into your car. Never in control of it.
We know "plug this into your car" is a big ask. Here's exactly what the Reachy dongle can and can't do.
how we ship
No update reaches you untested.
Safety rules are checked automatically
Before any update goes out, automatic tests make sure our safety rules still hold. If one fails, the update doesn't ship.
Tested on real phones
Every release runs on small and large iPhones and Android phones before anyone gets it.
Apple and Google review every version
No secret updates. Every public version of the app passes Apple and Google review first.
Nothing changed by hand
Our security settings are written down and reviewed, so nothing gets switched off by accident.
Changes tried on a copy first
Changes to how information is stored are tested on a copy before they touch the real thing.
Only the access that's needed
Every part of Reachy — and every person on our team — gets only the access their job needs.
what's next
Getting stronger before launch.
What's already done, and what we're adding.
For security researchers: the technical details+
- All clients call our own API gateway at the edge; no backend address or key ships in any app or page.
- Gateway allow-list → database grants (row-level security, no table access, explicit function allow-list) → per-function ownership checks.
- Rate limits per visitor, stricter on sign-in and code endpoints; request size limits.
- HTTPS only, TLS 1.2+ (1.3 on), HSTS, nosniff and no framing on every response. No certificate pinning by design.
- Logs carry route, status, timing and a request ID — never bodies, IPs, phone numbers, emails or locations.
- Secrets in an encrypted vault, injected at deploy, rotated at least yearly. Infrastructure as code, reviewed.
- Planned: App Attest / Play Integrity, durable audit log, CSP, dependency and secret scanning, external penetration test, SOC 2 Type I.
Responsible disclosure
Found a problem? Tell us first.
If you think you've found a security issue, email us the details. We read every report, answer quickly, and won't take action against honest research that respects our users' privacy.