CyberSnap AI

AI connects the evidence. You regain control.

During an attack, findings pile up faster than people can read them. CyberSnap AI turns the results of the cyber engine into a clear picture: what was found, what it means in your environment, which systems are affected and what to check next. It explains and recommends. Your team decides.

Available In the product today In development Not yet released to customers

Findings are not answers.

A scan of dozens of servers across weeks of recovery points can produce a long list of findings. In the middle of an incident, the team has to know which ones matter, whether they are part of the attack or normal activity, and what they mean for the return to operations. That is the work CyberSnap AI takes on.

01

Too much to read

Hundreds of findings across servers and points in time, each with its own file, rule and severity.

02

Little context

A finding on its own does not say whether it is the attack, a leftover from it or a legitimate tool.

03

Different audiences

Security, infrastructure and management need the same facts, explained at a different depth.

Six questions the team asks during an incident.

From the scan results screen, Ask AI produces six reports. Each one answers a question the incident team actually asks, based on the findings and the scan history.

Summary ReportAvailable

A management view of the scan: how many recovery points and servers were analyzed, whether risky files were found, which servers are at risk and the overall threat level.

Validate ThreatsAvailable

Reviews the findings one by one and assesses which look like real threats and which are more likely false positives, with the reasoning for each.

Show Affected ComponentsAvailable

Maps which servers, folders and recovery points are affected, so the team can see the scope of the incident at a glance.

Suggest Next ActionsAvailable

Turns the findings into a prioritized list of next steps: what to investigate, which earlier point to examine and which systems to check first.

Detect AnomaliesAvailable

Looks across the scan history for changes that stand out, such as a sudden jump in modified files or findings that appear on several servers at once.

Analyze Threat DetailsAvailable

A deep view of a specific threat: the detections behind it, including YARA rule matches, where it was found and what is known about it.

The Ask AI reports are released in the product as a beta. The User Guide on the Resources page describes each report.

How it works

Two engines. Clear roles.

CyberSnap runs two engines side by side. The deterministic cyber engine inspects the content of the recovery points and produces findings that are consistent and traceable: the same content, rules and engine version give the same result.

The AI engine works on top of those findings. It adds context, connects findings across servers and time, and explains what they mean. It can recommend that a finding be investigated further or treated as lower priority, but it never deletes a finding and never changes what the evidence says.

Cyber engineInspects file content and produces the findings.
AI engineExplains, connects and recommends, based on the findings.
Your teamReviews the reasoning and makes the decision.

Findings and context. Not your files.

It receives
  • Findings and their metadata: detection method, file names and paths, severity, number of occurrences and scan history.
  • Context about each asset, such as the server role, operating system and services.
It does not receive
  • Raw file content. Files are inspected by the cyber engine inside your environment.
  • Authority to act. The AI does not delete findings, change policies or restore systems.
It returns
  • A reasoned interpretation, a false-positive assessment and a severity recommendation.
  • Directions for further investigation, and what is still missing to reach a firm conclusion.

One set of facts, explained for each audience.

Security

Where to dig

Which findings look real, which look like noise, and what to investigate next.

Infrastructure

What is affected

Which servers and folders show findings, and in which recovery points.

Management

Where we stand

A plain summary of the incident and the options for returning to work.

Partners

What to tell the customer

A clear report an MSP or integrator can walk the customer through.

▸ WHAT COMES NEXT

From explaining findings to guiding the investigation.

01The AI proposes

Identifies what is missing and drafts a follow-up check.

02The cyber engine validates

Against allowed actions, scope and the approved run budget.

03The cyber engine runs it

Only what passed the check. New findings come back as evidence.

04The picture updates

The hypothesis is strengthened, weakened or narrowed.

Hypothesis-driven investigation is In development. The cyber engine stays in charge of what actually runs.

What teams ask about the AI.

Can the AI hide a finding?

No. The AI can recommend that a finding is likely a false positive, with its reasoning. The finding stays in the results, and your team decides how to treat it.

Does the AI decide which point to restore?

No. It recommends and explains. The people responsible for recovery make the decision, based on the findings and the business needs.

What happens without the AI?

The cyber engine and SnapMap work on their own. The AI adds context and speed in reading the results; it is not required to see the findings.

Does the AI keep learning?

Detection content and analysis are improved through CyberSnap AI Research, under a controlled release process with testing and customer approval where needed.

See the AI reports on your own findings.

Start with one system that matters to your business. Scan its existing snapshots and review the AI reports with our team.