ZX HUBUnified analysis server · enterprise · public sector

One high-speed
analysis server, shared
by every security layer.

ZX HUB is a Hyper Sandbox server that centrally analyzes files from multiple layers.
It finds unknown threats and blocks, quarantines and disarms them together with your existing security systems.

MTA · REST API · ICAP integrationRedundant configurationOn-premises · cloud
THREAT MAP

Threats no longer arrive by just one route.

AI-enhanced UNKNOWN malware targets email, network, cross-domain transfer and endpoints all at once.
It slips past spam filters and firewalls and spreads across both internal and internet-facing networks. The later it's detected, the bigger the damage.

Email in particular is the most dangerous route, accounting for 91% of successful cyberattacks.(Source: U.S. CISA)

ZX PLATFORM

Four layers,
watched on one screen

Files coming in through email, network, cross-domain transfer and endpoints are all inspected by a single analysis engine.
Unknown threats are isolated the moment they're found.

  • Threat detected
  • Isolated
Threat names shown are examples.
Email
Network
Cross-domain
Endpoint
PRODUCT

What's in ZX HUB

Hyper Sandbox, ZX CDR and the web console in a single server.
Click HYPER SANDBOX to analyze a file right away.

ZX HUB
Unified analysis serverFirewall · email · separated network · endpoint unified analysis · CDRIntegrates with security systems via ICAP / REST API / MTA
Components
  • HYPER SANDBOXISOLATED · ZX CORERuns unknown threats and judges them by behavior · 0.27 s on average
  • ZX CDRBehavior-basedPrecision disarm solution
  • URL ANALYSISURL behavior analysisL0–L3 multi-layer scanning · forced hyperlink execution
  • WEB CONSOLEStatistics dashboard · security policies · API management
0.27 sAverage analysis time per filePer KBeT testing
99%Detection rate for recent malware250,000 MalwareBazaar samples from 2022 onward
0%False-positive rateValidated on 260,000 clean files
4Security layers one ZX HUB coversEmail · network · cross-domain · endpoint
Analysis

Known threats fast, unknown threats deep

A pre-filter screens out known threats first; only the remaining files go to the Hyper Sandbox for deep analysis.
AI scores the risk and hands it to SOAR · SIEM as report.json.

File types HWP · DOC · XLS · PPT · PDF · ONE · VBS · JS · HTML · HTA · LNK · archives More on the technology →

ZX CDR

Remove only the threat, deliver the document intact

Only macros, scripts, URLs and hyperlinks in the mail body and attachments are removed. Mail is delivered as usual. Documents aren't blocked outright, so users don't complain.

  • MTA integration on the email layer
  • Analyze, block and disarm in one pass
  • Disarm report showing what was removed
ZX CDRBehavior-based precision disarmRemove only the threats; deliver the document intact.
  1. Original email · attachments
    MacrosScriptsURLsHyperlinks

    Threat elements hidden in the body and attachments

    Email (MTA) · network-separation integration
  2. ZX CDRThreat removal

    Behavior-based precision content disarm

    Behavior analysis and disarm in one pass
  3. Disarmed documentDelivered to the recipient as normal
    MacrosScriptsURLsHyperlinks

    Work continues with the cleaned file

    Needed documents arrive intact
Web console

From analysis to response in one flow

Handle file analysis reports and security policies in one flow in the web console.
Review verdict evidence, MITRE ATT&CK and IOCs, and set policies per file type.

Overview
Sample

Overview

See files, email, CDR, devices and system status on one screen.

As of 15:40
Malicious files and email foundIn the last 7 days we found 418 malicious files and 192 malicious emails. The latest was on Nov 12, 14:02.
Last 14 days 12 days with threats
3112345678910111213
Threats found No threats No scans
File scan
45,729scanned+10%
Malicious 418 Suspicious 125 Safe 45,186

2 devices sent files · 10,579 unique

Mail scan
1,582scanned+92%
Malicious 192 Other 1,390

Scans attachments and URLs in the body

Mail CDR
236processed+53%
Threats removed 144 No change 92 Other 0

Threats removed 78 · Repaired 88

Device
5registered
Connected 3 No response 10+ min 2 No records 0

2 devices have not responded for 10+ min

Scan trendsDaily · Each row has its own scale (see ‘Max’ at the top right of each row).
File scan45,729
Max 21K
Mail scan1,582
Max 480
CDR236
Max 56
Threats610
Max 268
11/711/811/911/1011/1111/1211/13
File riskVerdicts for scanned files.
99%Safe
  • Malicious4180.9%
  • Suspicious1250.3%
  • Safe45,18699%
Countries URLs point toShows which countries' servers the URLs in scanned files connect to.
South Korea 82,775USA 50,695France 6,045Ireland 1,754Japan 1,283Singapore 1,194Germany 1,047Netherlands 322UK 203Switzerland 116
  1. 1South Korea82,775
  2. 2USA50,695
  3. 3France6,045
  4. 4Ireland1,754
  5. 5Japan1,283
  6. 6Singapore1,194
  7. 7Germany1,047
  8. 8Netherlands322
  9. 9UK203
  10. 10Switzerland116

1 · 10 · 100 · 1K · 10K+ · URLs with unknown country 1,793 · 26 countries · Top 10 total 145,434

Top detected behaviorsRecorded even when not malicious.
  • FileEmbedded6,677
  • HyperLink2,023
  • TextMappedURL1,583
  • JSFound1,477
  • ArchiveLocked541
  • UnmatchedXref361
  • QRFound278
  • ObfuscatedPdf267
  • TailInjectedScript195
  • FakeEXT190
Device healthLast response time and protection status for each device.
● Connected 3● No response 10+ min 2Outdated version 0Real-time protection off 0
Device · UserLast seenVersionReal-time protection
mini-10.0.0.2419 days ago2026.11.3
mini-10.0.0.101User A · Dev team2 days ago2026.11.10
mini-10.0.0.165User B · Dev team8 min ago2026.11.11
mini-10.0.0.118User C · Dev team6 min ago2026.11.11
mini-10.0.0.180User D · Marketing3 min ago2026.11.11
Top scanned file types
  • HWP12,73028%
  • PDF12,48127%
  • DOC7,34216%
  • XLS4,3159.4%
  • XLSX1,9504.3%
  • Other6,91115%
System statusWhat scanning needs to run smoothly.
  • Data collectionFile, mail, CDR and device records all normal
  • Alert emailOn alert@example.com
  • UpdateAutomatic updates on
  • License48 days left until 2026-12-31
TypeNameVerdictDetectionTime
Filegrant_notice_sample.pdf3 duplicatesMaliciousDownloadFileExec11/12 14:02
Filegrant_application_sample.pdf3 duplicatesMaliciousDownloadFileExec11/12 14:02
Filebank_account_copy_sample.pdf2 duplicatesMaliciousDownloadFileExec11/12 13:51
Fileemployment_insurance_sample.pdf2 duplicatesMaliciousDownloadFileExec11/12 13:51
Fileemployment_insurance2_sample.pdf2 duplicatesMaliciousDownloadFileExec11/12 13:51
Filenational_pension_sample.pdf2 duplicatesMaliciousDownloadFileExec11/12 13:51
More on the ZX web console →

Keep your existing investment, add only an analysis layer

Plug in through standard integrations; no separate sandbox for each layer.
Mail and file transfers aren't delayed, so work keeps moving.

01 · SIMPLICITY

No technical barriers

Analysis that catches even evasive threats, plugged straight in via MTA · ICAP · REST API.

02 · EFFICIENCY

Cost-efficient

No per-layer sandboxes. $0 in OS and Office licenses for analysis, starting from a 4-core server.

03 · CONTINUITY

Business continuity

Mail and file transfers don't slow down. Add ZXONE Agent for ransomware blocking and recovery.

Frequently asked questions

Do we have to replace our existing security equipment?

No. ZX HUB integrates with what you have now, such as email security, NGFW, UTM and cross-domain solutions. It only adds unknown-threat analysis.

Will analysis slow down our work?

The average is 0.27 s per file (per KBeT testing). Even with hold-until-verdict as the default, users barely notice any wait.

What are the detection and false-positive rates?

We detected 99% of 250,000 MalwareBazaar samples from 2022 onward. The false-positive rate was 0% in validation on 260,000 clean files (over-detection cases found during validation were fed back into the engine).

Do we need a separate sandbox for each layer?

No. A single ZX HUB analyzes four layers: email, network, cross-domain and endpoint (integration with each layer required).

What does it cost, and how is it deployed?

From the KRW 10 million range, depending on the number of users (U100–U5000+). Delivered and installed as an all-in-one server, or as a cloud subscription.

Can we test it before buying?

Cloud PoCs can start right away. On-premises PoCs are arranged separately. We also run side-by-side demos with your own samples.

Is it certified?

TTA GS Certification Grade 1 · KBeT V&V test certification · selected as a 2025 KISA excellent information security technology · 12 patents. CC certification is under review.

Compare with your own samples

We demo side by side with your current solution, using suspicious files you already have.