Threats Tagged 'base64 encoding'

Xinference, an open-source distributed AI model inference framework, suffered a supply chain attack when attackers compromised PyPI release credentials of maintainers and published three malicious versions (2.6.0, 2.6.1, 2.6.2) on April 22, 2026. The malicious code, encoded in Base64 layers within __init__.py, executes automatically upon library installation or import, collecting cloud credentials, SSH keys, API tokens, database passwords, cryptocurrency wallets, and environment variables. The payload specifically targets AWS environments through metadata service exploitation and uploads stolen data to attacker-controlled infrastructure. The attack affects users who downloaded these versions from PyPI, which has over 680,000 total downloads. Attribution remains unclear as TeamPCP's name appears in the code but the group denies involvement, suggesting third-party impersonation.

TeamPCP launched a sophisticated attack on the Telnyx Python SDK, publishing malicious versions 4.87.1 and 4.87.2 to PyPI. The attack represents an evolution from their previous LiteLLM campaign, incorporating WAV-based steganography, split-file code injection, and expanded platform support. The payload, activated on import, uses stealthy techniques to download and execute credential-stealing malware across Linux, macOS, and Windows systems. Key changes include the use of audio steganography to hide malicious code, improved evasion through split-file injection, and the addition of Windows support with Startup folder persistence. The attackers shifted from HTTPS to plaintext HTTP infrastructure, potentially exposing their activities to network monitoring. Organizations are advised to downgrade to the last clean version and treat affected systems as compromised.

MediumCampaignUnited StatesGermanyUnited Kingdom+7#base64 encoding#pypi#persistence

This report details a multi-stage malware campaign delivering the REMCOS backdoor via a malicious Windows LNK shortcut file. The attack begins with social engineering, leveraging PowerShell for initial execution and deploys a persistent backdoor capable of full system compromise. The infection chain involves file download, Base64 decoding, and execution of a malicious PIF file masquerading as a Chrome-related program. The LNK file contains a PowerShell command that downloads and executes a payload, which is then decoded and run as CHROME.PIF. This file is identified as the REMCOS backdoor, capable of various malicious activities including keylogging, screen capture, and remote access. The attack utilizes multiple stages to evade detection and establish persistence on the victim's system.

MediumMalwareRomaniaGermanyFrance+4#base64 encoding#multi-stage attack#remcos