Infostealer Malware: Inside a Stealer Log and How to Respond
Skip to content

Infostealer Response

An infostealer is malware that copies credentials and session data from an infected device and sends them to an attacker. One infection can expose every password saved in the browser, along with the signed-in sessions that let an attacker skip multi-factor authentication.

What the attacker receives is called a stealer log. This guide opens one up, shows how its contents turn into an intrusion, and sets out which credentials to deal with first.

Inside a Stealer Log

A stealer log is a folder of files taken from one device. Each part answers a different question for the criminal who buys it.

Part of the logWhat it holdsWhat the buyer does with it
PasswordsEvery login saved in each browser: the site address, the username, and the password in clear text.Searches for corporate addresses such as a VPN portal, single sign-on page, or admin console.
CookiesSession cookies for every site the user was signed in to.Loads them into a browser to resume the session, with no password or second factor.
AutofillNames, addresses, phone numbers, and sometimes payment card details.Builds an identity for fraud or for answering security questions.
System informationComputer name, Windows username, operating system, IP address, country, installed software, and security products.Works out who the victim is, where they work, and whether the device is corporate.
ScreenshotThe desktop at the moment of infection.Confirms what the user was doing and which applications were open.
FilesDocuments from the desktop and common folders, often filtered by name or type.Looks for password lists, keys, and configuration files.
Application dataSaved credentials from VPN clients, remote desktop tools, file transfer programs, and chat applications.Direct access to the systems those tools connect to.

The malware runs for seconds. It collects, uploads, and in many cases removes itself. The user notices nothing, and by the time anyone looks there may be no file left to find.

The log is more dangerous than a list of leaked passwords. The site address is recorded with each credential, so the buyer knows exactly where it works. The cookies are current, so a session can be resumed before the password is ever needed.

The Infostealer Economy

Infostealers are a product. The people who write them do not usually infect anyone, and the people who use the stolen data did not usually steal it. Each step is a separate trade.

Developers

They write and maintain the malware and rent it out by subscription. Customers receive a builder, a control panel, and regular updates that keep pace with browser changes and security software.

Distributors

They rent the malware and spread it through search ads for fake software, poisoned downloads, links in video descriptions and forums, and fake verification pages. They are paid by the volume and quality of logs they bring in.

Log sellers

They run the markets and subscription channels where logs are sold. A buyer can search millions of logs for a company’s domain or a specific web address.

Access brokers

They buy logs in bulk, test the corporate credentials in them, and resell working access with a description of the victim.

End users of the access

These are ransomware affiliates, fraud groups, and intruders looking for a first foothold.

The economics explain the scale. A subscription costs little, distribution is automated, and one infected computer yields hundreds of credentials. Individual logs sell cheaply because there are so many. The value is extracted later, by whoever finds a corporate login among the personal ones.

A Documented Case

The 2022 breach of the Australian health insurer Medibank shows the full path from one saved password to a national incident.

In civil penalty proceedings filed in June 2024, the Office of the Australian Information Commissioner alleges that the breach began with the credentials of a contractor who worked on the IT service desk. According to the allegations, the contractor had saved work credentials in a personal browser profile, the profile synchronized them to a home computer, and malware on that home computer stole them.

The regulator further alleges that the attacker used those credentials to sign in to the company’s VPN, which did not require multi-factor authentication, and went on to take approximately 520 gigabytes of data. The proceedings concern the personal information of 9.7 million people. The case is before the Federal Court, and the allegations have not been decided.

Whatever the court finds, the sequence alleged is the standard infostealer intrusion:

  1. A work password saved in a browser.
  2. Browser synchronization carrying it to a device the organization does not manage.
  3. A consumer malware infection on that device.
  4. A remote access service that accepted the password alone.

No part of it involved an attack on the company’s own computers until the attacker signed in. Every step in the sequence has a control that would have broken the chain, and the Reduce the Risk section sets them out.

Why Infostealers Get Past Defenses

Infostealers are simple programs, and their simplicity is part of why they succeed.

They need no special privileges

A browser’s saved passwords and cookies belong to the user. Any program the user runs can reach them. The malware never needs administrator rights.

Most run once

A typical stealer installs no service and leaves nothing running, so a scan the next day finds a clean machine. Some arrive with a loader that stays behind and fetches more malware, which is why an infected device is untrusted until it is rebuilt.

The user starts them

The victim opens the file deliberately, believing it is an installer, a game modification, or a document. Some campaigns ask the user to paste a command to complete a fake verification step, which sidesteps download protections entirely.

They change constantly

Each customer receives a freshly built and packed copy, so file signatures are of limited use.

Browser protections are an arms race

Browsers encrypt saved data and keep adding safeguards. Stealer developers publish workarounds within weeks, because their business depends on it.

They live where corporate tools are absent

The home computer, the family laptop, and the contractor’s own machine have consumer antivirus at best.

The practical conclusion is that prevention on the endpoint will never be complete. The dependable defenses are the ones that make stolen data less useful: sessions that expire quickly, tokens bound to a device, and access that requires a managed machine.

From Infection to Intrusion

The trades described earlier work in sequence, and the gaps between the steps are where a defender has time.

1
Infection and uploadThe user runs the file, and within seconds the log is with the stealer operator.
2
SaleLogs are sold individually on criminal markets, or in bulk through subscription channels that publish thousands a day.
3
ValidationA buyer tests the corporate credentials and cookies found in the log. Working access to a business is worth far more than a personal account.
4
Resale or useValidated access is sold again, often to a ransomware affiliate, or used directly.
5
IntrusionThe attacker signs in as the employee. What follows is account takeover, data theft, or ransomware.

The timing depends on what was stolen. Cookies are tried within hours, while the sessions are still valid. Passwords circulate for weeks or months and are tried long after the infection was cleaned up. An exposure alert that arrives days after the infection is late for the sessions and early for the passwords, and both need action the day it arrives.

Which Device Was It?

The system information in the log identifies the infected machine, which decides how the organization can respond.

A managed work device

The computer name matches the organization’s naming scheme, so the security team can isolate it, collect evidence, and reimage it.

A personal device

The computer name is a home or default name, and the installed software includes games or consumer applications. The organization has no agent on it and no authority over it. The work credentials were there because the user signed in to work systems from home.

A contractor or supplier device

The username and installed software point to another company. The credentials are for the organization’s systems, and the cleanup belongs to someone else.

Exposure alerts for personal and contractor devices arrive with nothing from the organization’s own endpoint protection, because that protection was never on the device.

Detecting an Infection on a Managed Device

Where the infected computer is one the organization manages, endpoint telemetry can show the theft as it happens. The behaviors are distinctive.

A program reading browser data files

Browsers store saved logins and cookies in known files inside the user’s profile. A process other than the browser opening them, particularly for several browsers in a row, is the core signal.

Execution from the Downloads folder or a temporary folder

A newly downloaded program, or one unpacked from an archive, runs within minutes of arriving.

A script engine started from a pasted command

PowerShell or a similar tool is launched by the user with a long encoded command, immediately after browsing.

An archive created in a temporary folder and uploaded

The collected data is usually compressed before it is sent.

One outbound connection to an unfamiliar address

A single upload goes to a host nothing else in the organization contacts, sometimes to a messaging platform’s interface.

A screenshot taken by a non-interactive process

This is part of the standard collection routine.

The program deleting itself

The file removes itself shortly after first running.

Detection that waits for a human to read an alert will usually be reviewing history, which is still valuable. It identifies the device, the time, and the user, and starts the response before the log is sold.

When a managed device is confirmed infected, treat every credential the user had saved as stolen, even if no exposure alert has arrived yet.

Triage by Credential Type

A single log can list hundreds of credentials. Deal with them in order of the access they give.

PriorityCredentialWhyAction
1Single sign-on and identity provider logins, with their cookiesOne account opens every connected application, and a live cookie bypasses the second factor.Reset the password, revoke sessions, review authentication methods.
2VPN and remote access portalsA direct route into the internal network.Reset, check for a second factor, review recent connections.
3Administration consoles: cloud, firewall, hypervisor, backup, remote managementControl of infrastructure.Reset, rotate any keys, review administrative actions.
4EmailFraud, further phishing, and password resets for other services.Reset, revoke sessions, check rules and forwarding.
5Code repositories and developer toolsSource code, and secrets stored in it.Reset, rotate tokens and keys, review access logs.
6Business applicationsCustomer, finance, and HR data.Reset, review activity.
7Personal accountsOut of the organization’s control, though reused passwords matter.Advise the user to change them.

An exposure alert usually reports only the entries that matched the organization’s domain. The same log holds the user’s logins for suppliers, cloud consoles, and personal accounts, and those are equally known to the buyer.

Response Order

The order matters. A new password typed on a device that still carries malware goes straight to the attacker.

1
Identify the deviceUse the computer name and username in the log, and confirm with the user which devices they use for work.
2
Stop using the device for workIsolate a managed device. Ask the user to stop using a personal one.
3
Block sign-in where the risk is highIf the account is privileged, or the sign-in logs already show use from elsewhere, block it before anything else.
4
Reset passwords from a clean deviceWork in triage order. Do this before revoking sessions, or the attacker signs straight back in with the saved password.
5
Revoke sessionsStart with the identity provider and the other critical systems at the top of the triage table. This is the step that ends the stolen cookies.
6
Remove what an attacker addedRecord each item, then remove any authentication method, registered device, rule, or application grant the user does not recognize.
7
Rotate other secretsThis covers API keys, access tokens, and SSH keys stored on the device.
8
Confirm in the sign-in logsReview each account’s activity since the infection date, and check that nothing from unfamiliar sources succeeds after the reset.
9
Rebuild the deviceReimage a managed device. A personal device needs a clean reinstall before it is trusted with work again.

The account takeover guide explains the account steps, and the session hijacking guide shows what each action ends.

Was the Access Used?

The infection date in the log sets the start of the period to review. For each exposed account, look for the following.

Session use from elsewhere

An existing session shows activity from an address the user has never used, with no sign-in recorded there.

Sign-ins with the stolen password

Successful sign-ins arrive from hosting providers or other countries, or attempts are stopped by the second factor.

Remote access connections

VPN or remote desktop sessions come from unfamiliar addresses.

Changes that secure access

New authentication methods, registered devices, application grants, or mailbox rules appear.

Administrative actions

New accounts, role changes, or configuration changes are made with the exposed account.

Other infected devices

The same download source or file turns up elsewhere in the fleet.

Where any of these appear, the incident is an account takeover or a network intrusion and should be handled as one.

A Policy for Personal Devices

Most infostealer exposure comes from devices the organization does not manage. A clear position on personal devices does more than any amount of endpoint tooling on the corporate fleet.

The workable positions, in order of strictness, are these.

1
Managed devices onlyAccess to work systems requires a device the organization controls and can verify. Sign-ins from anything else are blocked. This removes the problem and is practical where everyone has a work laptop and phone.
2
Browser-only access from personal devicesStaff may read mail and documents in a browser on a personal device. Downloading, synchronizing, and saving are blocked, and sessions are short. A stolen cookie expires quickly and no files reach the device.
3
A managed work profileOn phones and some computers, a separate managed container holds work applications and data, with its own controls, while the rest of the device stays personal.
4
Unrestricted with monitoringAny device may sign in. The organization relies on detecting misuse. This is the default in many organizations and the position that produces most exposure alerts.

Whichever is chosen, a few rules help.

Keep work passwords out of personal browsers

Tell staff never to save them there, and never to sign a personal browser profile in on a work computer. Browser synchronization quietly copies saved passwords to every device using that profile.

Cover contractors and suppliers

Their staff sign in to the organization’s systems from devices nobody has assessed. Contract terms can require managed devices or restricted access.

Decide what happens after an exposure

Staff should know that a personal device found infected has to be cleaned before it is used for work again, and who will help them do it.

Reduce the Risk

The first controls here follow the sequence in the documented case. The rest limit what a stolen log is worth.

Stop browsers saving work passwords

Provide a managed password manager, and switch off the browser’s own password store through browser policy on work devices.

Stop synchronization to personal profiles

Browser policy on work computers blocks sign-in to personal profiles, so saved work passwords are not copied to a home device.

Require a second factor on every remote access service

A VPN, remote desktop gateway, or support tool that accepts a password alone turns one saved password into network access.

Replace passwords where possible

A passkey gives a stealer no password to copy.

Bind sessions to devices

Token protection and compliant-device policies stop a stolen cookie or token being used from another machine.

Shorten sessions on sensitive systems

A stolen cookie for an admin console should expire in hours.

Control installation

Restrict what users can install and which browser extensions are allowed.

Monitor exposure

Watch stealer logs and breach data for the organization’s domains, and treat each alert as an incident with a deadline.

The browser and installation controls reach only managed computers, which is why the position taken on personal devices matters as much as any setting in this list.

Infostealer Response With ThreatDefence

ThreatDefence monitors leaked and exposed credentials for customer domains. When staff logins appear in stolen data, our analysts turn the exposure into a prioritized list of password resets and session revocations, and check sign-in records for use of the stolen access.

Our endpoint agent collects forensic evidence from an infected device, and our SOC can revoke sessions and disable accounts under agreed authority. Our security advisories track current campaigns, including infostealers and session theft.

A Microsoft 365 forensic scan in the ThreatDefence portal with a replayed session token finding, a new MFA method, an app consent and the sessions revoked in response

Experiencing a breach?

Our incident response team answers 24x7, with or without a retainer.

Talk to Our Experts

Ask about credential exposure monitoring and a free Microsoft 365 forensic review for your organization.

    I'm an MSP, MSSP, or reseller

    By submitting this form, you accept the terms in our privacy policy.

    Frequently Asked Questions

    What is an infostealer?

    An infostealer is malware built to copy credentials from a computer and leave. In seconds it collects the passwords and cookies saved in every browser, autofill data, files from common folders, and logins stored by VPN and remote access clients, then uploads them as one package. It is rented as a subscription service, and well-known families include Lumma, RedLine, and Vidar.

    What is a stealer log?

    A stealer log is the package one infection produces: a folder for one device holding its saved passwords with the site each belongs to, browser cookies, autofill data, system details, a screenshot, and selected files. Logs are sold singly or by subscription, and buyers search them by web address. One log commonly holds hundreds of logins, most of them personal and a few of them corporate.

    Can an infostealer bypass multi-factor authentication?

    Yes, through the cookies. A session cookie is issued after the user has passed the second factor, and a service that receives the cookie from another browser treats the session as already authenticated. The stolen password alone is still stopped by MFA. For an exposed single sign-on account the sessions have to be revoked as well as the password reset, and short session lifetimes and device binding limit what a stolen cookie is worth.

    Is changing the password enough after an infostealer infection?

    No. The log also holds live session cookies, which on many services keep working after a password change until the sessions are revoked, and it may hold API keys, SSH keys, and tokens stored on the device. A new password typed on the infected device can be stolen again. Reset from a clean device, revoke sessions, rotate the other secrets, and rebuild the computer before it is used for work.

    How do we find out if staff credentials are in stealer logs?

    Credential exposure monitoring searches stealer logs and breach data for the organization’s email domains and for the web addresses of its sign-in pages, VPN, and admin consoles. Searching by address matters, because it finds contractors and personal accounts that sign in to company systems. National cyber agencies also notify organizations of exposures they come across, and Microsoft Entra ID flags users whose credentials appear in leaks.

    What should we do if an employee's personal device was infected?

    Treat every work login saved on it as stolen. Have the employee stop using the device for work, reset their work passwords from a clean device, revoke their sessions, and review the accounts for sign-ins since the infection date. The organization cannot examine or reimage a personal computer, so the condition for using it again is a clean reinstall, and the employee needs help changing their personal passwords too.

    How is a stealer log different from a data breach list?

    A breach list comes from one compromised service and holds that service’s usernames and passwords, often hashed and often years old. A stealer log comes from one person’s computer and holds their logins for every service, in clear text, with the exact sign-in address and current session cookies. A breach list feeds credential stuffing. A stealer log gives direct, targeted access.

    Do infostealers lead to ransomware?

    Often. Access brokers buy logs, test the VPN, remote desktop, and administrator logins in them, and sell working access to ransomware affiliates. The Australian Signals Directorate has warned that stealer infections on personal devices are being used to compromise corporate networks. A remote access service that accepts a password alone is the usual way in, and the gap between the sale of the log and the intrusion can be days or months.