Zynap recognized by Gartner® as a Sample Vendor for Preemptive Exposure Management in the Gartner Agentic Security Report. Read more!

Security Automation

What Is Malware Sandboxing? A Guide for Security Teams and MSSPs 

Some malware stays quiet until a computer has been switched on for hours. The PhantomRAT sample our Labs team documented in August waits about 2.7 hours, so it can look harmless in a freshly started sandbox. Here's how malware sandboxing works, and what a sandbox report should tell your team before they act.

Author

default avatar

Zynap Team

What Is Malware Sandboxing? A Guide for Security Teams and MSSPs 

Malware sandboxing is a cybersecurity method for running a suspicious file in an isolated environment and recording what it does. Where a signature scan checks a file against threats that are already known, a sandbox lets the file run and watches its behavior.

The sandbox turns what it sees into a report that opens with a verdict of malicious, suspicious or clean. That verdict is a useful first answer, but on its own it won’t tell you what the file did or which domains and IP addresses to block. A good report answers those questions too, so a first-line analyst can close the case without passing it to a specialist.

This guide explains how malware sandboxing works, how it compares with antivirus and Endpoint Detection and Response (EDR) tools, and why some malware behaves differently inside a sandbox. It also covers what to look for in a sandbox report and who can see the files you upload.

Key Takeaways on Malware Sandboxing

  • Malware sandboxing runs a suspicious file in an isolated environment and records what it does.
  • Static analysis examines a file without running it, while dynamic analysis watches what it does when it runs.
  • A sandbox can show what new malware does even when no antivirus signature exists for it yet.
  • Some malware checks whether it’s inside a sandbox and hides its harmful behavior if it finds signs of one.
  • A good sandbox report explains what the file is, what it did and which indicators to act on.

What Is Sandboxing in Cybersecurity?

Sandboxing in cybersecurity means running software in a restricted environment that’s kept apart from the rest of the system. A malware analysis sandbox uses that isolation to detonate suspicious files and record their behavior as evidence. Those files can arrive automatically from email security gateways and EDR tools, and your analysts can upload any file they’d like to check.

How Does Malware Sandboxing Work?

Malware sandboxing works by examining a suspicious file before it runs, then running it in an isolated environment and recording its behavior. Security teams call these two steps static analysis and dynamic analysis, and each one finds things the other can’t.

Static Before Dynamic

Static analysis comes first because it’s quick and the file never runs. The sandbox reads what’s inside the file, like the sections of a Windows program, the macros in an Office document and any web addresses written into it. For example, static analysis of a Word invoice can show a macro set to run when the document opens.

Static analysis can’t read code that’s packed or encrypted, and many attackers hide their code this way. That code only appears in memory once the file runs, so the sandbox moves on to dynamic analysis.

What Gets Recorded During Detonation

Detonation means running the file so the sandbox can watch it, and most sandboxes do this inside a virtual machine that’s isolated from your network. While the file runs, a good sandbox records:

  • The processes the file starts, like a Word document starting PowerShell
  • The Application Programming Interface (API) calls it makes to the operating system, with their arguments
  • The domains it looks up and the IP addresses it connects to
  • The files and registry keys it creates, changes or deletes
  • The code it unpacks into memory while it runs
  • What appears on screen, as screenshots or a video

Arguments matter because they show exactly what happened. A call on its own tells you the file changed the registry, while its arguments show which key it changed and the value it wrote. In the invoice example, they’d show the web address the macro contacted and the name of the file it saved. Our engineering post explains how the sandbox records every API call with its arguments if you’d like the technical detail.

Sandbox vs Antivirus vs EDR

Antivirus, EDR and a malware sandbox all help you deal with malicious files, but each one does a different job. If you’re already running antivirus and EDR, the useful question is what a sandbox adds.

Antivirus protects your devices from threats that are already known. It’s fast and good at blocking common malware, but a new or disguised file can still get past it.

EDR watches activity on your laptops and servers and alerts your team when something looks wrong. Its job is to stop an attack early on a real device, so it won’t always show you everything a file was built to do.

A malware sandbox answers the question both of them can leave open. It tells you what a suspicious file is, what it did and what to block, even when other tools don’t recognize it.

What Is Sandbox Evasion?

Sandbox evasion is when malware looks for signs that it’s running in a sandbox and hides its harmful behavior if it finds them. When it works, the sandbox can return a clean verdict for a file that’s dangerous.

MITRE ATT&CK groups the checks evasive malware uses into three kinds:

  • System checks, like the size of the hard drive or software that only virtual machines run
  • User activity checks, like mouse movement or the number of files on the desktop
  • Time-based checks, like how long the computer has been switched on

For example, the PhantomRAT malware our Labs team documented in August only continues once the computer has been on for about 2.7 hours. Many sandbox machines are started just before the test, so a sample like this can stay quiet and look harmless.

A sandbox catches more evasive malware when it looks and behaves like a real computer. That takes monitoring from below the malware’s own code, where it’s harder to spot, and realistic answers when the malware asks about the hardware. In the Zynap Sandbox, you can also choose the country a sample detonates from, because some malware only acts when it thinks it’s in a certain place. Our Labs team explains the engineering in its post on kernel-level malware analysis from Ring 0. If you’re comparing sandboxes, ask how each one handles evasive malware before you trust a clean verdict.

How to Use a Sandbox Report for Malware Triage

Malware triage is the first look at a suspicious file, when your team decides how dangerous it is and what to do next. A good sandbox report supports it in four steps, in the order you’ll need them. The Zynap Sandbox report is built around the same four steps.

Check the Verdict and Threat Score

The verdict tells you how dangerous the file looks, and many sandboxes back it with a threat score. Together they’re the quickest way to sort a busy queue, so start with them.

Identify the Malware Family

A malware family is a group of malware samples that share the same code and behavior, like LockBit ransomware or the Lumma infostealer. A sandbox can’t always name the family, but when it does, you know what the file was built to do and what’s already known about it.

Review the Behavior and MITRE ATT&CK Mapping

The behavior section shows what the file tried to do, like survive a restart, hide inside other processes, avoid analysis or reach the internet. Mapping each behavior to MITRE ATT&CK gives it a standard name your team and tools already use. A plain-language summary and a screen recording synced to the events let a first-line analyst follow what happened. Deep work on the code itself, like reverse engineering, still goes to a specialist.

Collect the Indicators of Compromise (IOCs)

Indicators of Compromise (IOCs) are the traces a malicious file leaves behind, like domains, IP addresses, file hashes and registry keys. The report should list them all in one place, so your team can block them or search for them across your network.

What Is IOC Extraction?

IOC extraction pulls Indicators of Compromise out of a sandbox report in a format your other security tools can use. An IOC sitting in a PDF doesn’t protect anything, and that’s why this hand-off is a natural place for cybersecurity automation. Extracted IOCs can go to four places.

IOC Blocking

Firewalls, email security gateways and EDR tools can block the domains, IP addresses and file hashes the sandbox found. The same file then can’t reach other devices or talk to its servers.

Automated Hunting Workflows

Hunting workflows search your laptops and servers for the same IOCs, so you can find other devices the file reached before it was blocked. With security workflow automation, they can start as soon as a new report is ready.

IOC Matching in Your SIEM

Your Security Information and Event Management (SIEM) platform can match the IOCs against the logs it already holds. A match shows you where they’ve already appeared, even before anyone knew they were malicious.

YARA Rules for Malware Detection

A YARA rule describes text and code patterns that a malware family’s files share, so your tools can spot related samples even when their file hashes change. Your team can write one from the sandbox’s findings, and some sandboxes can produce one from the malware’s own code.

Public vs Private Malware Sandboxes

The main difference between a public and a private malware sandbox is who can see the files you upload. Many public sandboxes are free to use, but the files and reports from free accounts can be visible to other users.

That matters during an incident for two reasons. A suspicious file can hold your own data, like an invoice with a customer’s details, so a public upload can expose data you’re required to protect. Attackers can also watch public services for their own files, and a public upload can tell them you’ve spotted the attack.

A private sandbox keeps new samples inside your own tenant, the separate space your organization has in the platform. Files that have already appeared in public feeds are a different case, because they’re already visible to anyone.

Malware Sandboxing With Zynap

The Zynap Sandbox is the malware analysis sandbox in our preemptive security automation platform, and it’s live for every Zynap customer. Each file is detonated in isolation, recorded in full and read back as a report, and that analysis feeds what the platform does next.

On Windows 10, the run is monitored from the kernel and a modified hypervisor, so more evasive malware shows its behavior. One analysis can then serve several teams at once.

For MSSPs

When it’s done by hand, malware analysis for every client is slow and needs specialists. With the Zynap Sandbox, Managed Security Service Providers (MSSPs) can offer it to every client, without the workload growing as fast as the client list. Here’s what that can look like:

  • Close triage at first line, using the narrative and the recording.
  • Keep new samples private in your tenant.
  • Act on each client’s IOCs in their own tools, with per-tenant workflows.
  • Use the plain-language narrative in your client reporting, with no rewrite from a specialist.
  • Set up the same service for new clients quickly, with reusable workflows.

For Security Operations Teams

When a file is detonated for up to five minutes, its full report is ready about 5 to 20 minutes after it’s submitted. Each report gives your first-line analysts what they need to decide a case on their own:

  • A verdict and a score out of 100, so they can sort a busy queue at a glance.
  • A chaptered narrative of what the file did, with every claim linked to its evidence.
  • A recording of the run next to a live event feed, so they can watch the invoice’s macro start PowerShell at the second it happened.
  • A summary of thousands of events, with suspicious and malicious patterns called out by name.

When the answer is clear, they can document the case and close it without escalating. The IOCs it extracts go straight into the blocking and hunting workflows already running in the platform.

For Threat Intelligence Teams

Every detonation adds to the intelligence your team owns:

  • The malware family and its behavioral tags, with classification and certificates to support attribution.
  • The malware’s extracted configuration, so you can see the infrastructure behind it and where to hunt next.
  • A network summary and a world map of every location the sample contacted.
  • In advanced mode on Windows 10, the CPU context of every API call, the runtime strings and a YARA rule produced from the code you select.

A private sample and its analysis stay in your tenant, next to the intelligence in our threat intelligence platform.

For Offensive Security Teams

Red and purple teams can seed their exercises with the behavior of real malware. The report shows the real chain of execution, process by process, so an exercise can test your defenses against what a sample did. Preparation scripts let you set up the machine before the sample runs, so the run matches the environment you’re testing.

How to Get Started With the Zynap Sandbox

If you’re already a Zynap customer, the Sandbox is ready in your platform. You can start with a single file the next time something looks suspicious, then read the report and see which IOCs it extracts for your workflows. Once you trust what it finds, add the Sandbox to a workflow, so suspicious files from your other tools reach it automatically.

Learn how the Zynap Sandbox can support your malware triage by booking a demo.

Frequently Asked Questions

What Is a Malware Sandbox?

A malware sandbox is an isolated environment where security teams run a suspicious file on purpose and record what it does. The result is a report that shows what the file is, what it did and which indicators to act on.

What Are the Different Types of Sandboxing?

The main types are browser sandboxing, application sandboxing and malware analysis sandboxes. Browsers and operating systems use sandboxing to limit what websites and apps can reach. Security teams use malware analysis sandboxes to run suspicious files on purpose and study what they do. Email and web security tools also sandbox attachments and links, so they're checked before your users open them.

What's the Difference Between a Sandbox and a Virtual Machine?

A virtual machine is a simulated computer, and most malware sandboxes use one as the place where a file runs. The sandbox adds what a plain virtual machine doesn't have. That includes tools that record what the file does, controls that keep it away from your network and a report that explains the results.

Is There a Free Sandbox for Malware Analysis?

Yes, several public services analyze files for free. The trade-off is visibility, because files and reports from free accounts can be visible to other users. For incident material or regulated data, a private sandbox is the safer choice.

Is a Sandbox 100% Safe?

No sandbox is 100% safe, and neither is any other security tool. Evasive malware can stay quiet inside it, and in rare cases attackers have found flaws that let code break out of a virtual machine. That's why a sandbox needs regular updates and controlled network access.

Is Windows Sandbox Safe for Malware Testing?

Windows Sandbox is built for running untrusted apps in a disposable desktop, but it isn't a malware analysis tool. Microsoft notes that its networking is enabled by default, and that this can expose untrusted apps to your internal network. Out of the box, it also doesn't record what a file does or give you a report, so you'd only see what appears on screen.

The verdict is only where a sandbox report starts. What earns a sandbox its place is how quickly your team can act on everything after it.

Book a demo and we'll show you around

Close your exposure window and stay ahead

By clicking the button above, I consent to Zynap, storing and processing the personal information submitted above to provide me the content requested in accordance with the Privacy Policy. In compliance with the information obligation established by the data protection regulation, we provide you the information regarding the processing of your personal data, how to unsubscribe, as well as our privacy practices and commitment to protecting your privacy in our Privacy Policy.