This blog post was authored by Edmund Brumaghin and Holger Unterbrink with contributions from Eric Kuhla and Lilia Gonzalez Medina.


Cisco Talos has recently observed multiple campaigns using the Remcos remote access tool (RAT) that is offered for sale by a company called Breaking Security. While the company says it will only sell the software for legitimate uses as described in comments in response to the article here and will revoke the licenses for users not following their EULA, the sale of the RAT gives attackers everything they need to establish and run a potentially illegal botnet.

Remcos' prices per license range from €58 to €389. Breaking Security also offers customers the ability to pay for the RAT using a variety of digital currencies. This RAT can be used to fully control and monitor any Windows operating system, from Windows XP and all versions thereafter, including server editions.

In addition to Remcos, Breaking Security is also offering Octopus Protector, a cryptor designed to allow malicious software to bypass detection by anti-malware products by encrypting the software on the disk. A YouTube video available on the Breaking Security channel demonstrates the tool's ability to facilitate the bypass of several antivirus protections. Additional products offered by this company include a keylogger, which can be used to record and send the keystrokes made on an infected system, a mass mailer that can be used to send large volumes of spam emails, and a DynDNS service that can be leveraged for post-compromise command and control (C2) communications. These tools, when combined with Remcos provide all the tools and infrastructure needed to build and maintain a botnet.

Within Cisco's Advanced Malware Protection (AMP) telemetry, we have observed several instances of attempts to install this RAT on various endpoints. As described below, we have also seen multiple malware campaigns distributing Remcos, with many of these campaigns using different methods to avoid detection. To help people who became victims of a harmful use of Remcos, Talos is providing a decoder script that can extract the C2 server addresses and other information from the Remcos binary. Please see the Technical Details section below for more information.

Technical Details

Remcos distribution in the wild

Talos has observed several malware campaigns attempting to spread Remcos to various victims. Since Remcos is advertised and sold on numerous hacking-related forums, we believe it is likely that multiple unrelated actors are leveraging this malware in their attacks using a variety of different methods to infect systems. Earlier this year, RiskIQ published a report regarding an attacker who was reportedly targeting defense contractors in Turkey. Since then, this threat actor has continued to operate and has been observed targeting specific types of organizations. Talos has confirmed that in addition to defense contractors, this attacker has also targeted other organizations such as:

  • International news agencies;
  • Diesel equipment manufacturers and service providers operating within the maritime and energy sector; and
  • HVAC service providers operating within the energy sector. In all of the observed campaigns, the attack begins with specially crafted spear phishing emails written in Turkish. The emails appear as if they were sent from a Turkish government agency and purport to be related to tax reporting for the victim's organization. Below is an example of one of these email messages:

The attacker put effort into making the emails look as if they were official communications from Gelir İdaresi Başkanlığı (GIB), the Turkish Revenue Administration, which operates under the Ministry of Finance and is responsible for handling taxation functions in Turkey. The attacker even went as far to include official GIB graphics and the text at the bottom which translates to:

"Thank you for your participation in the e-mail notification system of [the] Department of Revenue Administration's e-mail service. This message has been sent to you by GIB Mail Notification System. Please do not reply to this message."

As is common with many spear phishing campaigns, malicious Microsoft Office documents are attached to the emails. While the majority of these documents have been Excel spreadsheets, we have also observed the same attacker leveraging Word documents. In many cases, the contents of the document have been intentionally blurred as way to entice victims to enable macros and view the content. Below is an example of a Word document associated with one of these campaigns that have been made to look as if it is a tax bill:

Many of the Excel spreadsheets we analyzed were mostly blank, and only included the following image and warning prompting the victim to enable macros in Turkish:

We have also observed campaigns that appear to be targeting English-speaking victims. Below is an example of one of the malicious attachments that were made to appear as if it was an invoice on letterhead associated with Iberia, which is the flagship airline in Spain.

In addition to the Iberia-themed malicious documents, we uncovered multiple malicious documents that were created to appear as if they were invoices associated with AMC Aviation, a Polish charter airline. Talos has observed the following same itinerary decoy image used across both Excel and Word documents:

As described in the RiskIQ report, the macros in these files contain a small executable that is embedded into the document in the form of a series of arrays. When executed, the macros reconstruct the executable, save it to a specific location on the system and execute it. The file location specified changes across malicious documents, but includes directories commonly used by malware authors such as %APPDATA% and %TEMP%. The executable filename also changes across documents.

The extracted executable is simple and functions as the downloader for the Remcos malware. It is a very basic program and is used to retrieve Remcos from an attacker-controlled server and execute it, thus infecting the system. An example of this is below:

Remcos is a robust RAT that can be used to monitor keystrokes, take remote screen captures, manage files, execute commands on infected systems and more. In several cases, the distribution servers associated with these campaigns have been observed hosting several other malicious binaries in addition to Remcos.

Who is behind Remcos?

As previously mentioned, a company called Breaking Security has been offering Remcos and other questionable software for purchase on their website. There are no details about the company or the people behind it listed on its website. The website does, however, list a value-added tax (VAT) number (DE308884780) which shows the company is registered in Germany. Interestingly, you can look up the name and address of companies in almost any European Union (EU) country except Germany on this website. Germany does not share this information due to privacy concerns. Because Breaking Security was registered in Germany, we were unable to identify the name and address of the individual behind this company. Nevertheless, we were able to identify several artifacts that give us an idea as to who might be behind the company.

Comparison of Public and Private VAT Entries
The Breaking Security domain is hosted behind Cloudflare currently, and Whois privacy protects the registrant information. Quite a bit of effort has been put into attempting to mask who is behind this company and the associated software. During our analysis, we were able to uncover several clues about the individual that we believe is behind this organization, either due to mistakes or very well organized false evidence on the internet.

The first thing we identified was the following email address and domain present in the Viotto Keylogger screenshot below:


While the viotto-security[.]net domain server and registrant information is protected similar to what was seen with the breaking-security[.]net domain, the domain viotto[.]it listed in the "Sender's e-mail" text field is not. The Whois information associated with this domain can be seen in the screenshot below:

Normally Talos would obfuscate this data however since it is public in so many places we have elected not to. We also identified additional email, Jabber, and XMPP addresses that appear to be used by the author of Remcos by leveraging the data we collected from the website, as well as other sources:


In multiple cases, the domains investigated were leveraging the Cloudflare service. This often obscures the address of servers hosting domains, as the DNS configuration typically points the name resolution to Cloudflare IPs rather than the IP of the web servers themselves. One common mistake is that while the domain itself may be protected by Cloudflare, in many cases, a subdomain exists that does not point to Cloudflare servers, allowing the server IP address to be unmasked.

This was the case with the breaking-security[.]net domain. While Cloudflare shields the domain, their mail subdomains are not protected. The A record that was configured for the mail subdomains is as follows:

mail[.]breaking-security[.]net. A 146.66.84[.]79
webmail[.]breaking-security[.]net A 146.66.84[.]79

The IP address 146.66.84[.]79 is hosted at SiteGround Amsterdam. After various testing, we are confident that this is also the IP address where the main breaking-security[.]net website is hosted.

One of the other domains we identified as being associated with Remcos was viotto-security[.]net. This domain is currently configured to redirect traffic to the main breaking-security[.]net domain. However, this was not always the case. Searching for pages associated with this domain in the Wayback Machine, a website that allows users to view past versions of a web page, yields the following result in the form of a personal biography. There are multiple clear overlaps between the interests of this individual and the developer of the various tools the company sells:

We also identified several instances where Viotto was advertising, selling and supporting Remcos on various hacking forums, including HackForums since at least 2016, which makes their intentions questionable. Below is an example of one of these threads.

While the company states that they revoke user licenses if they were to use Remcos for illegal activity, as illustrated by the thread below the purported official reseller of Remcos doesn't seem to mind another user informing it that they are using the software to control 200 bots.

Viotto also appears to be active on other hacking forums, including OpenSC, where he is a moderator. Below is a thread where this user is advertising Remcos and Octopus Protector.

Remcos Technical Details:

As described in other blog posts, Remcos appears to be developed in C++.

As the release notes show, it is actively maintained. The authors release new versions on almost a monthly basis:

v2.0.5 – July 14, 2018
v2.0.4 – April 6, 2018
v2.0.3 – March 29, 2018
v2.0.1 – Feb. 10, 2018
v2.0.0 – Feb. 2, 2018
v1.9.9 – Dec. 17, 2017

Remcos has the functionalities that are typical of a RAT. It is capable of hiding in the system and using malware techniques that make it difficult for the typical user to detect the existence of Remcos.

Several routines are looking like they were just copied and (best case) slightly modified from publicly available sources. A good example is the anti-analysis section:

It is checking for an outdated artifact, the 'SbieDll.dll'. In our opinion, there are not many analysts using Sandboxie these days anymore. A closer look at the other functions is also showing a high code similarity to publicly available projects. Below you can see the Remcos VMware detection code:

The following is a code sample from

The blog referenced above has already described several functions of Remcos features in detail. We would like to focus on Remcos' cryptographic implementation. It uses RC4 pretty much everywhere when there is a need to decode or encode any data. Examples are registry entries, C2 server network communication or file paths shown below:

The exepath registry data is base64-encoded, RC4-encrypted data. Decoded, it is the path of the executable:

The RC4 implementation is the standard RC4 implementation that can be found in many code examples on the internet. They are first building the Key Scheduling Algorithms (KSA) S_array at 00402F01.

This can be converted into the typical RC4 pseudo code:

for i from 0 to 255
        S[i] := i
j := 0
for i from 0 to 255
        j := (j + S[i] + key[i mod keylength]) mod 256
        swap values of S[i] and S[j]

Which is followed by the RC4 Pseudo-random generation algorithm (PRGA) at 00402F5B.

Which looks in pseudo code like this:

i := 0
j := 0
while GeneratingOutput:
        i := (i + 1) mod 256
        j := (j + S[i]) mod 256
        swap values of S[i] and S[j]
        K := S[(S[i] + S[j]) mod 256]
        output K

As the screenshots above illustrate, Remcos is using RC4 to encrypt and decrypt its data, and it is using the PE resource section to store the initial encryption key in the 'SETTINGS' resource. This key can have a variable length — we have seen short keys from 40 bytes to keys with more than 250 bytes.

They are storing the data in the following format:

[Length of key]
[Encryption Key]
[Encrypted configuration data]

This encrypted configuration data section contains the command and control servers, RAT commands to execute and other data. Decoded, it looks like this:

The decoded data contains the C2 server, e.g. ejiroprecious[.]ddns[.]net, and the corresponding port number, followed by a password. This password is used to generate a separate S_array for the RC4 encrypted C2 communication. The picture shows the relevant part of the RC4 Key Scheduling Algorithms (KSA) from above.

Even if a stronger password is used than in the example above, using such a weak encryption algorithm means that everyone who gets his or her hands on the binary file can extract the password and decrypt the C2 traffic or inject their own commands into the C2 channel to control the RAT. The good news is that companies who became a victim of Remcos have a good chance to analyse the threat if they have stored the network traffic and the Remcos binary file.

To make the life of forensic investigators easier, we are providing a small decoder Python script that can decode the config data from the resource section:

As mentioned above, Remcos is using the same encryption routine for all kinds of other functions, too. For this reason, the decoder program also offers an option to hand over encrypted bytes manually. This can be used to decode, for example, the exepath registry key.

We have used this tool to extract all the IOCs below. It is tested with the latest 2.0.4 and 2.0.5 versions of Remcos, but likely also works with other versions.

The user can also copy bytes from a network sniffer to a binary file, and hand it over to decrypt the bytes from the C2 communication to see which commands the C2 server has sent to the victim. Keep in mind to use the extracted password, e.g. "pass."


While the organization that sells Remcos claims that the application is only for legal use, our research indicates it is still being used extensively by malicious attackers, as well. In some cases, attackers are strategically targeting victims to attempt to gain access to organizations that operate as part of the supply chain for various critical infrastructure sectors. Organizations should ensure that they are implementing security controls to combat Remcos, as well as other threats that are being used in the wild. Remcos is a robust tool that is being actively developed to include new functionality increasing what the attackers can gain access to. To combat this, organizations should continue to be aware of this threat, as well as others like this that may be circulated on the internet.


Additional ways our customers can detect and block this threat are listed below.

Advanced Malware Protection (AMP) is ideally suited to prevent the execution of the malware used by these threat actors.

Cisco Cloud Web Security (CWS) or Web Security Appliance (WSA) web scanning prevents access to malicious websites and detects malware used in these attacks.

Email Security can block malicious emails sent by threat actors as part of their campaign.

Network Security appliances such as Next-Generation Firewall (NGFW), Next-Generation Intrusion Prevention System (NGIPS), andMeraki MX can detect malicious activity associated with this threat.

AMP Threat Grid helps identify malicious binaries and build protection into all Cisco Security products.

Umbrella, our secure internet gateway (SIG), blocks users from connecting to malicious domains, IPs, and URLs, whether users are on or off the corporate network.

Open Source Snort Subscriber Rule Set customers can stay up to date by downloading the latest rule pack available for purchase on

Indicators of Compromise (IOC)

The following IOCs are associated with various malware distribution campaigns that were observed during analysis of Remcos activity.

Malicious Office Documents:


Stage 1 Executables:


PE32 Executables:

A text file containing a list of Remcos PE32 executable hashes can be found here.

IP Addresses: