Backup Recovery Assurance Agent
Finds out what would not come back
- 8hard rules
- Things it must never do, enforced in every run
- 5handover rules
- Conditions where it stops and gives the work to a person
- 11output fields
- A structured result, not a paragraph you have to re-read
- ≤7steps per run
- Bounded work: at most 5 tool calls, enforced by the runtime
Read-only by design. This agent has no path to sending, spending, publishing or committing anything. Where that limit is the product, removing it would remove the reason to trust the output.
Overview
Almost every organisation backs things up and almost none can say what would come back. The job reports success, the dashboard is green, and everyone treats that as the answer. A successful backup job proves a file was written. It proves nothing about whether the data is complete, whether it can be read, or how long it would take to make the business work again.
The failures are found in a crisis, which is why the story is always the same. A system nobody added to scope because it was created after the last review. Backups running successfully to a location the same event would destroy. Retention shorter than the time it takes to notice corruption, so by the time you know, the clean copy has aged out. A recovery time nobody has measured against one the business has promised customers.
And the one that ends companies: the backups are reachable with the same credentials as the systems they protect, so whoever reaches the primary reaches the copies.
This agent reviews the arrangement from your records and configuration and reports what would not come back, ranked by what it would stop. It restores nothing and tests nothing — a restore is a live operation on real systems, and it belongs to the people who own them.
Benefits
You find out before the incident
Which is the only useful time to find out.
The gap between promise and reality
Named, with what it would take to close it.
Evidence for auditors and customers
Beyond a screenshot of a green dashboard.
What it does
Finds what is not backed up at all
Usually a system created after the last scope review. Nobody decided to exclude it.
Checks the copies are isolated
Backups reachable with the same credentials as the systems they protect are not backups. This is the one that ends companies.
Tests retention against detection time
If corruption takes six weeks to notice and retention is thirty days, the clean copy is already gone.
Compares measured recovery to promised recovery
Against what you told customers and auditors, not against what the tool claims.
Distinguishes a job from a restore
A green dashboard proves a file was written. Nothing more, and everyone reads it as everything.
Restores nothing
A restore is a live operation on real systems. It reviews the arrangement and reports.
What this changes
What this changes about your week
- It takes over
- Establishing what would actually come back after a failure.
- Done today by
- Assumed from a green dashboard.
- How often
- Annually, on paper.
- Typical manual effort
- Coverage, isolation, retention and tested recovery times read against what was promised.
Why IT & Service Management in particular. A successful backup job proves a file was written and nothing else. And backups reachable with the same credentials as the systems they protect are inside the blast radius rather than the recovery from it.
Runs unattended
Started by you or by an event, and it finishes on its own. Nothing waits for someone to be at a desk.
The same standard every time
The two-hundredth item is held to the bar the first one was. Consistency is the part people cannot sustain.
It cannot act on its own
Backup Recovery Assurance Agent has no path to sending, spending or committing. That limit is why its output is safe to act on.
How it runs
This agent runs server-side through the PROMIVO runtime. Each run is logged step by step and every tool call is permission-checked before it executes.
- 1InputValidated against the agent input schema
- 2Modelclaude-opus-5 with automatic fallback
- 3Toolssearch_knowledge, extract_structured_data
- 4PermissionsRuns READ actions automatically; anything higher waits for approval
- 5OutputStructured result matching the output schema
- 6LoggingTokens, duration and every step recorded
Demo
Demo dataIllustrative sample output, abridged.
Input
{
"systems": [],
"asAtDate": "2026-09-02",
"typicalDetectionDays": 45
}Output
{
"escalate": true,
"disclaimer": "A review of the configuration and records supplied. No restore, backup, deletion or failover has been performed or simulated, no system was accessed, and no statement is made that any backup is valid or restorable — only what has and has not been tested.",
"notBackedUp": [
"The document signing service, added in March, is not in backup scope. Nobody decided to exclude it — it was created after the last scope review."
],
"objectiveGaps": [
{
"stated": 4,
"status": "unevidenced",
"system": "Customer database",
"measured": "Never tested"
},
{
"stated": 24,
"status": "missed",
"system": "Reporting warehouse",
"measured": "31 hours in the last test"
}
],
"couldNotAssess": [
"Backup completeness — no job-level record of what was included was supplied."
],
"rankedByImpact": [
"Backup isolation — affects everything, and is the difference between an incident and an extinction event.",
"Customer database has never been restore-tested.",
"Retention shorter than detection time."
],
"escalationReason": "Backups share credentials with the systems they protect, the primary customer database has never been restore-tested, and retention is shorter than the stated detection time.",
"isolationFindings": [
{
"issue": "Backups are written to storage reachable with the same administrative credentials as the systems they protect.",
"system": "All production systems",
"consequence": "Anyone who reaches production reaches the copies. In a ransomware scenario the backups are part of the blast radius rather than the recovery from it."
}
],
"retentionFindings": [
"Retention is 30 days. Your own stated typical detection time for data corruption is 45 days. By the time you know, the clean copy has aged out."
],
"neverRestoreTested": [
"The primary customer database has never had a restore tested. Every backup job for it has reported success for three years, which proves a file was written and nothing else."
],
"sharedFateWithPrimary": [
"Three systems back up to the same cloud region as their primary. A regional failure removes both at once."
]
}Requirements
Integrations
No integrations required.
Before you start
- Systems in scope, with their business criticality
- Backup configuration: schedule, retention, destination, credentials model
- Restore test history, if any
- Your stated recovery time and recovery point objectives
- Requires the All-Access plan or higher
Use cases
Annual continuity review
What would actually come back, and how fast.
Before a certification audit
The questions an auditor asks, answered first.
After adding systems
Whether backup scope followed them.
Ransomware preparedness
Whether the copies survive an attacker who reaches the primary.
Pricing
All-Access
Billed monthly through your PROMIVO subscription. Cancel at any time.
Runs consume your plan allowance for agent executions and tokens. See plan limits
FAQ
Does it perform test restores?
No. A restore is a live operation on real systems with real consequences. It reviews configuration, records and test history, and reports what has never been tested — which is usually the finding.
Does it connect to our backup system?
No. It works from exports and documentation you supply, so it needs no access to anything that could be used to alter or delete a backup.
What if we have never tested a restore?
Then that is the report's first line. An untested backup is a belief, and saying so plainly is more useful than any other finding it could produce.
Reviews
Reviews open once customers have run this agent.
Need this agent to work differently?
Tell us what to change and our team will scope a customised version for your business.
Customize this agent