Security and platform
Security described by its mechanisms, each with a test behind it.
Each hosted company runs as its own instance with its own database. Staff log in with mandatory two-factor, every request is checked on the server, and money and audit records are append-only.
- Roles with default permissions plus a per-user switchboard; every change is audited.
- Deactivation takes effect on the next request; API keys go inert with the account and never need reissuing.
- Mandatory two-factor login for staff, trusted devices that can be revoked, rate limiting and lockouts, passwords hashed with bcrypt or argon2.
- Each hosted company is a full separate instance: its own database, files and domain. Isolation is structural, not a query filter.
- Outbound calls guarded against SSRF; uploads validated, with the size asked before the bytes are read; escaping at every document boundary.
- Append-only financial and audit records; encrypted credentials; encrypted off-site backups to a public key, with tested restores.
- A review programme in which every guard ships with a test that fails when the guard is removed.
A guard and its test
- Two-factor login for stafffails if the enforcement is removed
- Permission check on every requestfails if a door forgets it
- Append-only ledgerfails if an update gets through
- SSRF guard on outbound callsfails if a private address passes
Every guard ships with a test that fails when the guard is removed.
Want to see the system on cases from your own work?
Write to us on WhatsApp or by email and we will arrange a walkthrough on examples from your operation.