Security
Oria is in beta. We treat security reports seriously and reply quickly. This page tells you how to report something, what to expect from us, and what's in scope.
Why I built Oria
I built Oria because I was tired of losing track of my own life: subscriptions I forgot I had, bills that slipped, documents I could never find when I needed them. I wanted one calm place that quietly keeps everything and surfaces what matters, without selling me out.
I use Oria every day, and I hold your data to the standard I want for my own. It is encrypted, isolated per account, read-only where it touches your email, and deletable in one click. If something here worries you, write to me directly: hi@heyoria.com.
Issam, founder of Oria
How to report
Email security@heyoria.com. PGP is fine; ask if you want our key.
Include, at minimum:
- A clear description of the issue.
- Steps to reproduce. If your steps involve a test account, tell us which one.
- The impact you believe it has. We'll calibrate; you don't need to grade it for us.
- A timeline preference if you have one (e.g. plan to publish in 90 days).
What you can expect from us
- Acknowledgement within 72 hours. A human will reply.
- An honest assessment of severity and a target fix window.
- Credit in the release note if you want it. We will not name you without permission.
- No legal action against good-faith research that respects the scope below.
We are a beta product without a bug bounty program. We appreciate responsible disclosure and try to make the process feel respectful.
In scope
heyoria.comand any subdomain we operate.- The Oria mobile/desktop app, when those exist.
- The Supabase project we run (vulnerabilities specific to our configuration of it).
- The Python extraction sidecar we deploy on Railway.
Out of scope
- Third-party services we use (Supabase platform itself, Anthropic, Resend, Vercel, OpenAI). Please report those upstream.
- Social-engineering attacks against Oria employees or other users.
- Physical attacks against Oria infrastructure or staff.
- Findings that require already-elevated access (e.g. you already have the service-role key).
- Denial-of-service findings whose realistic exploit volume is what every public web service already accepts.
- Self-XSS, missing security headers without a demonstrated attack, and reports that copy automated-scanner output without analysis.
Safe harbour
If you act in good faith, stay within the scope above, and don't exfiltrate or destroy other people's data, we will not pursue legal action and will work with you on disclosure.
Machine-readable contact: /.well-known/security.txt.