Trust
Responsible disclosure
If you find something, tell us before you tell anybody else, and we will behave well about it.

How to report
Email security@algoryq.com with what you found, how to reproduce it, and the impact you think it has. If you need to send something sensitive, say so and we will arrange a way.
What we commit to
- Acknowledged within two working days, by a person, not an autoresponder.
- Our severity assessment, and why, within five working days.
- Updates while we fix it, rather than going quiet.
- Public credit if you want it, silence if you do not.
- No threats, and no lawyers, over a report made in good faith under this policy.
In scope
- The product at the application and API domains.
- This website.
- Anything that lets one institute reach another's data. This is the one we care about most: isolation is the load-bearing claim of the whole system.
- Authentication, session handling and permission bypasses.
Out of scope
- Findings from automated scanners with no demonstrated impact.
- Missing headers on a static page that has no session and sets no cookie.
- Social engineering of our people, or physical access.
- Denial of service. Please do not.
- Anything requiring a compromised device or an already-phished person.
What we ask while you look
- Use your own test institute. Do not touch anybody else's data.
- If you reach data that is not yours, stop and tell us what you saw so we can assess the exposure. Do not download it.
- Do not degrade the service for anybody else.
- Give us a reasonable window before publishing. We will not use it to stall.
Money
No bug-bounty programme: we have no revenue yet, and running one badly is worse than not running one. Better said plainly than implying a reward that does not exist. When there is a programme, this page will say so.
The controls you would be testing
Row-level isolation, deny-by-default authorization and a hash-chained audit log, with the file each one lives in.
Or open the sandbox (opens in a new tab) and click around first. No form.