Missing data is never good news
When an export carries no CVE references, many tools print an empty section that reads as a clean result. Scan Hub marks the CVE set Unknown and says why.
Scanner-agnostic · AI-written · Exploit-aware
Scan results in. Client‑ready report out.
Drop in an export from your network, host or web vulnerability scanner. Scan Hub ranks every finding by real-world danger and writes the fix plan your client will sign off on.
First three reports free, no card needed. The pre-flight check needs no account at all. See sample reports →
Clients
Open findings
412
Known exploited
9
Overdue
14
Fixed, last 30 days
63
| Client | Known exploited | Open | Last scan |
|---|---|---|---|
| Northwind Dental | 3 | 88 | 2 days ago |
| Harbor Logistics | 2 | 131 | 5 days ago |
| Elm Street Law | None | 24 | Yesterday |
- Tenable Nessus
- Qualys
- Rapid7
- Greenbone / OpenVAS
- Nuclei
- Microsoft Defender
- ConnectSecure
- Other vulnerability scanners
How it works
Three steps. No spreadsheets.
Upload the export
CSV, XML or JSON from your network, host or web vulnerability scanner. No agents and no scanner credentials. The file is checked before you pay, and anything unreadable is named.
AI reads and ranks it
Every finding is checked against CISA's Known Exploited Vulnerabilities catalog, exploit likelihood and vendor fixes, then grouped into the fewest actions that close the most risk.
Send, ticket and track
Hand the client a branded PDF, push the fixes to your PSA as tickets, and see what was fixed when the next scan arrives.
The report
Ranked by what attackers actually use.
Severity scores alone bury the dangerous items. Scan Hub puts known-exploited and ransomware-linked findings first, then the rest in order, with an executive summary in plain English and the technical detail your engineers need underneath.
Open the sample reports- Band 1Known exploited, used in ransomwareFix first. From the CISA KEV catalog.
- Band 2Known exploitedAttackers are using it now.
- Band 3Critical CVE, CVSS 9.0 or higherSevere if reached.
- Band 4CVE rated medium or abovePlanned patching.
- Band 5No CVE, rated medium or aboveConfiguration and hardening.
- Band 6Low and informationalRecorded, not urgent.
Evidence first
AI that won't tell your client a gap is a clean bill of health.
Conflicts are kept, not merged
When one scan gives two different CVE sets for the same check, the disagreement is evidence. Scan Hub quarantines the finding instead of picking a value.
Every sentence traces to evidence
The AI writes only from facts the rules engine extracted. Every number in the report comes from the scan or a named intelligence source.
Same scan, same plugin
- Variant A — CVE-2024-0001, CVE-2024-0002
- Variant B — CVE-2024-0009
Data-quality state: quarantined
Both variants are preserved and the conflict is recorded. Varaxon Scan Hub does not pick a winner, and the prioritized output marks the CVE set for this finding as Unknown until a human resolves it.
Supported scanners
Bring the export you already have, or connect directly.
Support is defined by the export, not the scanner edition. For anything not listed, the free pre-flight check tells you whether Scan Hub can read it, and the AI maps the columns when the report is written. If it can't, the credit is refunded.
Tenable Nessus
.nessus XML or CSV export
Tenable Nessus report guideQualys
XML or CSV scan report
Rapid7
Nexpose / InsightVM XML export
Greenbone / OpenVAS
Greenbone XML report
Greenbone / OpenVAS report guideNuclei
JSON or JSONL output
Microsoft Defender
Connected directly; needs Defender for Business (in Microsoft 365 Business Premium), Defender for Endpoint Plan 2, Defender Vulnerability Management or Defender for Servers
ConnectSecure
Connected directly with your ConnectSecure API key
Other vulnerability scanners
CSV, XML or JSON exports that list findings by host, IP or URL
Who it's for
For whoever has to explain the scan.
MSPs
A report per client per month without the copy-paste, with fixes pushed straight into your PSA.
MSSPs
Exploit-aware priorities and overdue tracking across every client you watch.
Security consultants
Deliverables in your branding that read like you wrote them, in a fraction of the time.
Home and small office
Ran a scan and can't make sense of it? Get a clear list of what to fix, and how.
Sample reports
Judge the output before you sign up.
Both samples are synthetic demonstration reports generated from test exports. They contain no customer data.
Tenable Nessus
Tenable Nessus report sample
.nessus XML export
Generated from a synthetic three-host Nessus export that carries CVE references. It opens with an executive summary and the decisions leadership needs to make, ranks the findings into priority bands using CISA KEV and EPSS data, then sets out a remediation plan and what the scan cannot show.
Synthetic demonstration report generated from a test export. Not a customer report.
Download sample PDF (Tenable Nessus)
Greenbone / OpenVAS
Greenbone / OpenVAS report sample
Greenbone Community Edition XML export
Generated from a Greenbone Community Edition scan of Metasploitable 2, a deliberately vulnerable lab machine. The export carries no structured CVE references, so the report says exploitation status is unknown and ranks findings by scanner severity instead. That is not the same as a clean scan.
Demonstration report generated from a lab scan. Not a customer report.
Download sample PDF (Greenbone / OpenVAS)Pricing
Per report, or per month.
Your first three reports are free, with no card and no automatic conversion.
Starter
$49 / month
or $490 / year
- 10 reports per month
- 1 user
Core
$99 / month
or $990 / year
- 25 reports per month
- Up to 3 team members
Growth
$199 / month
or $1,990 / year
- 60 reports per month
- Up to 10 team members
Scale
$299 / month
or $2,990 / year
- 150 reports per month
- Unlimited team members
Pay-as-you-go report: $9 per report, no subscription. Full pricing →
Security and data
Your clients' data, handled like it matters.
Uploads are used only to produce your report. They are never sold, never shared for advertising and never used to train AI models. Google Ads and Reddit Ads measurement load only if you accept advertising cookies.
MFA for everyone
Every Scan Hub account signs in with a second factor.
Names stay private
Host names, addresses and emails are swapped for placeholders before anything reaches the AI.
Named providers
Reports are written with Anthropic's Claude and stored with Supabase. Both are named in our privacy policy, and neither trains on your data.
Walled-off data
Each organisation's data is isolated from every other at the database level.
Try it on your last scan.
Run the free pre-flight check and see what your export supports.
