On October 21, 2026, the LiteLLM Python proxy library version 1.107.0 was compromised in a supply chain attack. The malicious release was available on PyPI and official Docker images for approximately two hours. Attackers used stolen GitHub CLI OAuth tokens from three engineers to deploy a malicious fork via a private release workflow. The payload included a credential stealer targeting cloud, developer, AI-tool, browser, and LiteLLM proxy credentials, as well as a second-stage remote access trojan backdoor. The malware executes automatically whenever Python starts by placing files in the site-packages directory. This is the second known supply chain compromise of LiteLLM in 2026, unrelated to the previous incident linked to TeamPCP activity.
A supply chain attack targeting the Microsoft DurableTask Python client compromised versions 1.4.1, 1.4.2, and 1.4.3 on PyPi. The threat actor gained access through a compromised GitHub account previously linked to attacks, using stolen credentials to dump GitHub secrets containing PyPi tokens. The evolved payload targets Linux systems, stealing credentials from AWS, Azure, GCP, Kubernetes, Vault, and password managers like Bitwarden and 1Password. It propagates via AWS SSM and Kubernetes lateral movement, limited to 5 targets per infected host. The payload scrapes shell history, bruteforces password managers, and establishes persistence through infection markers. Compromised packages were quarantined following analysis.
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.