GitHub payloads sharpen Python stealer risk

Cybersecurity teams are tracking a stealthy espionage-style malware operation that uses GitHub Releases to hide a Python-based infostealer behind humanitarian-themed phishing material aimed at Russian-speaking targets.

The campaign, tracked as Operation HumanitarianBait, begins with phishing emails carrying a RAR archive that contains a malicious Windows shortcut file. The lure presents itself as a Russian-language humanitarian aid request or application form, exploiting a sensitive social context while keeping the victim’s attention on a decoy document as the infection chain runs silently in the background.

Researchers assessing the operation found a multi-stage intrusion process built around a PE-less Python architecture, meaning the main implant avoids the conventional Windows executable format that many security products are tuned to inspect. The shortcut file is unusually large because it contains embedded, obfuscated Unicode content. PowerShell extracts that content from within the file, decodes it and executes it in memory, reducing the chance that automated sandbox systems will capture the full behaviour.

Once triggered, the malware creates a self-contained Python environment under the user’s AppData directory, using the name WindowsHelper to resemble a legitimate component. The installation does not require administrator privileges, making it suitable for attacks against standard user accounts in offices, charities, aid-linked networks and civil-administration environments. Silent VBScript launchers invoke pythonw. exe, allowing the implant to run without a visible console window.

GitHub Releases sits at the centre of the delivery strategy. Instead of relying only on suspicious attacker-controlled domains, the operator hosts payload components as release assets on a trusted developer platform. Clean components, including the Python embedded runtime and installer utilities, are placed alongside malicious files, creating traffic that can appear consistent with normal software updates. A repeatedly republished data archive with changing hashes indicates that the operator is actively modifying or recompiling the payload to weaken static detection.

Persistence is established through Windows Task Scheduler. A task named WindowsHelper is configured to run at short intervals and survive reboot, giving the operator long-term access after the first infection. This approach reflects a broader shift in malware tradecraft, where attackers seek durability without noisy system changes that would draw user attention or endpoint alerts.

The implant’s capabilities go beyond basic password theft. It targets credentials and cookies from Chromium-based browsers and Firefox, including stored secrets protected by newer browser encryption methods. It also records keystrokes, monitors clipboard contents, captures screenshots and searches user directories for high-value files such as documents, configuration files, source code, credential stores and strings resembling cryptocurrency private keys.

Telegram session folders are also targeted, raising the risk of account takeover without the need for passwords. By stealing session data, attackers may gain access to private chats, organisational channels and contact networks, making the campaign valuable for intelligence collection as well as financial theft.

Remote access capability further increases the threat. The malware can silently deploy legitimate remote desktop tools such as RustDesk and AnyDesk, hide their windows and transmit connection details to the command-and-control infrastructure. This gives the operator interactive control over infected systems while blending into software commonly used by support teams and administrators.

Observed infrastructure includes a command-and-control server used to deliver decoy material, receive stolen data and manage live access. The presence of more than one lure type, including a survey-themed variant, suggests the operator is testing delivery methods rather than relying on a single phishing template. The Russian-language material and targeting logic point towards Russian-speaking individuals or organisations, though firm attribution to a named threat actor has not been established.

Operation HumanitarianBait also underlines the growing abuse of trusted platforms in malware staging. GitHub, package registries, cloud storage services and collaboration tools are attractive to threat actors because many organisations allow them through firewalls and proxy filters. Blocking such platforms outright is often impractical, especially for software, research and engineering teams, creating a defensive gap that attackers increasingly exploit.



Notice an issue?

Arabian Post strives to deliver the most accurate and reliable information to its readers. If you believe you have identified an error or inconsistency in this article, please don't hesitate to contact our editorial team at editor[at]thearabianpost[dot]com. We are committed to promptly addressing any concerns and ensuring the highest level of journalistic integrity.


ADVERTISEMENT