Backups & recovery
We back up every ExoHuman Cloud instance. You do not need to configure anything.
What is backed up
Section titled “What is backed up”- The instance database — users, agents and their configuration, conversations, executions, schedules, projects, inbox items and the audit log.
- Agent workspaces — every agent’s home and workspace volume, including files and git history.
- Encrypted credentials and the keys needed to decrypt them, stored separately from the data they protect.
- Instance configuration — the settings that make your instance yours.
Schedule and retention
Section titled “Schedule and retention”Backups run daily and are kept for the retention period of your plan:
| Limit | Launch | Scale | Enterprise |
|---|---|---|---|
| Seats (people with an account) | 5 | Unlimited | Unlimited |
| Agents | 10 | Unlimited | Unlimited |
| Concurrent agent runs | 3 | 50 | By contract |
| Included storage | 50 GB | 500 GB | By contract |
| Backup retention | 7 days | 30 days | 90 days |
Scale and Enterprise also include point-in-time database restore, so the database can be restored to a specific moment within the retention period, not only to the last daily backup.
Backups are stored separately from the running instance, in the same region, and encrypted at rest.
Backups around updates
Section titled “Backups around updates”Before every ExoHuman update we take an additional backup of the database and instance configuration — and of agent volumes when the release changes the agent base image. If the update fails verification, it is rolled back automatically to that backup. See Uptime & SLA.
Restore tests
Section titled “Restore tests”We regularly restore backups into an isolated environment to prove they work, rather than discovering problems during an incident.
Requesting a restore
Section titled “Requesting a restore”Email [email protected] from an administrator account with:
- your instance URL;
- what to restore — the whole instance, the database to a point in time, or a single agent’s workspace;
- the point in time you need.
Whole-instance and database restores replace current data with the backup, so we confirm with you before starting and agree a window. Single-agent restores can usually be done without affecting other agents.
Your own copies
Section titled “Your own copies”Backups are a disaster-recovery tool, not an archive. For data you must keep independently:
- sync important agents to your own git repository (Memory & git state);
- export the audit log on a schedule;
- download reports and files you rely on.
See also Cancelling & data export.