Private laptop in the home office as a gateway for stealing browser passwords and sessions
Technical ArticlesT-46115 Jul 20267 min

DumpBrowserSecrets: when the home office becomes a security trap

Or: why hardened company devices do not protect against password theft from private PCs – and what organisations need to do about it.

A tool currently attracting attention in security circles highlights an often underestimated risk: DumpBrowserSecrets – a relatively simple program that extracts passwords, refresh tokens, session cookies, credit card data, autofill entries, browsing history and bookmarks from modern browsers (Chrome, Edge, Firefox, Opera, Vivaldi and others). What makes it stand out: it does not need admin rights.

At first glance, this may sound like a theoretical risk. But for organisations with hybrid or remote working models, it is a very concrete and frequently overlooked security gap – precisely because it does not affect the hardened company devices, but employees’ private computers at home.

The scenario: why private PCs become critical infrastructure

The typical hybrid working setup today: an employee works on a hardened company laptop with endpoint detection and response, mobile device management and conditional access. That machine is well protected, MFA is enforced, VPN is mandatory.

At the same time, the same employee works on their private PC at home in the evening or at the weekend – sometimes out of necessity, sometimes out of habit. And on that private machine, anything goes: unsafe software can be downloaded, old operating systems run unpatched, antivirus is optional, browser security settings are left at their defaults or not enabled at all.

This is where DumpBrowserSecrets comes in. An attacker – whether via a trojan, a malicious attachment in a phishing email or any other malware sample – only has to run this tool. No admin rights required. No suspicious permission prompts that would make the employee think twice. And suddenly the attacker has:

  • All saved passwords of the employee – often the same ones used on the company laptop or in other critical systems.
  • Active session cookies that let them sign straight into email, cloud storage, Git repositories and SaaS tools – without going through MFA, because the session has already been validated.
  • Refresh tokens that remain valid longer than session cookies and allow the attacker to keep access over a longer period.
  • Browsing history and bookmarks that often reveal sensitive information about internal services, APIs or tools.
  • Credit card data that can be used for identity theft or to buy hacking tools.

All of this data – usually unencrypted or protected only by weak, locally stored keys – can be extracted from a private PC in a matter of seconds.

Why this is so dangerous

From an attacker’s perspective, this is a jackpot scenario:

  1. Low barrier to execution: the tool requires no special technical skills. It can be bundled into a larger malware package or automated with a script.
  2. High success rate: on private PCs without standard security controls, it works reliably.
  3. Direct access to company resources: session cookies give immediate access to the same tools the employee uses – without cracking a password and without MFA raising an alarm.
  4. Window of opportunity: the organisation may only notice the attack days or weeks later – for example when suspicious activity appears from an unusual IP address, data is exfiltrated or altered, or another security control is triggered.

For attackers, an employee’s compromised private PC is often a better entry point into the corporate network than a direct attack on hardened company resources.

Also a standard red team tool

DumpBrowserSecrets is not only a tool misused by malicious actors. Red teams, penetration testers and security researchers also use it as a demonstration tool – to show exactly this weakness. We use similar concepts in our own penetration tests to make clients aware of these risks.

That does not change reality, though: once something is a red team tool, it will soon be (or already is) integrated into malware kits with malicious intent.

What organisations need to do

The good news: there are proven measures to make this attack vector visible and contain it. It does, however, require a multi-layered strategy.

1. Active Microsoft 365 monitoring and in-depth analysis

  • Anomaly detection in sign-in behaviour: if the same user suddenly signs in from a new IP address, at an unusual time of day or from a new device, audit logging should capture it.
  • Real-time alerts for suspicious activity: token refreshes from unexpected devices, mass file access, suspicious mail forwarding rules – all of this must be monitored.
  • Correlated analysis: a single suspicious sign-in may be a false positive. But if a session token has been stolen and several activities from unusual locations follow shortly afterwards, that is a clear signal.

Useful tools include Microsoft Entra ID Protection (sign-in risk detection), cloud access security brokers such as Microsoft Defender for Cloud Apps or Netskope, and threat intelligence feeds that track compromised session tokens or attacker IP addresses.

2. Harden conditional access policies

  • Reduce token lifetime: if a session token is valid for a maximum of 8 hours rather than 24 hours or more, the attacker’s window shrinks.
  • Refresh token rotation: a new refresh token should be issued every time one is used. This makes a stolen token worthless faster.
  • Enforce MFA for sensitive operations: even within an already validated session, certain actions (large downloads, adding users, permission changes) can trigger a fresh MFA prompt.
  • Check device compliance: if a device is identified as non-compliant (e.g. a private PC without security software), sensitive operations should be blocked – or at least require an additional authentication step.
  • Geographic anomalies: a token used in London that suddenly shows up in Bangkok is suspicious – even if the session is technically still valid.

3. Awareness and policies for employees

  • Define BYOD policies: who may work from private devices, and under what conditions? Often the answer is “not at all”; sometimes it is more pragmatic.
  • Train secure remote working: employees should understand that the security of their home office PC is just as critical as that of their company laptop – because both give access to the same resources.
  • Be careful with browser password storage: a password the browser remembers and you never type yourself is easily forgotten – and is therefore often not changed after it has been stolen.

4. Technical measures on private systems (where feasible)

  • Endpoint detection and response (EDR) on private devices too, if employees work from them – and enforced on all company devices at the very least.
  • Browser security policies: modern browsers offer options to disable or restrict password saving.
  • Hardware security keys and Windows Hello: authentication methods based on private keys that cannot easily be extracted.

5. Forensic readiness

When such an incident occurs, a fast response is critical:

  • Audit log analysis: which activities were carried out with the stolen session? From which IP addresses? When?
  • Browser forensics on the employee’s PC: was the tool actually present? When was it executed?
  • Token revocation: stolen tokens must be invalidated immediately.

At DFIR Austria, together with Probatio Cyber Defence, we carry out these analyses regularly – both forensically after a known incident, as part of a Microsoft 365 forensic investigation, and proactively in penetration tests and red team exercises.

Conclusion: a new reality of hybrid work

The shift to hybrid work and BYOD has fundamentally changed organisations’ attack surface. From a security standpoint, an employee’s own hardware is no longer “private” – as soon as that employee accesses company resources, their PC becomes part of the IT infrastructure.

Tools like DumpBrowserSecrets make it clear that hardening company devices alone is not enough. The real security test is what happens when a private PC is compromised – and how quickly it is detected. For organisations, this means:

  1. Monitoring is not optional – active Microsoft 365 monitoring and anomaly detection are essential.
  2. Conditional access policies must be genuinely conditional – not just “MFA is on”, but intelligent rules that understand context.
  3. Build forensic capabilities – when an incident occurs, you need to establish quickly what actually happened.
  4. Employees are part of the solution – awareness and clear policies matter as much as technology.

This complexity is also why managed security operations (SOC) and professional penetration testing are becoming ever more important. The attack surface has become too large and too dynamic to cover with rule sets alone – it takes continuous monitoring, continuous testing and an experienced team that interprets the signals correctly.

Sources