Malware · 3h ago
Wiz says malicious LiteLLM 1.107.0 was uploaded to PyPI on October 21 and appeared in Docker images tagged v1.107.0, latest, main-latest and main-stable. The package stole browser, cloud, developer and AI-provider credentials. Wiz reported that PyPI quarantined the release and the affected image tags were removed that day; 1.106.2 was the latest safe version.
The package installs files that Python automatically runs when an interpreter starts. That means credential theft can trigger even when a program never imports LiteLLM. Wiz also found a dormant loader that could fetch and run more code, though its remote instruction file was empty during analysis.
The exposure reaches beyond LiteLLM proxy deployments: if a CI runner or container installed 1.107.0 and then launched Python, secrets available there may already have left the environment. Removing the package or image ends the startup trigger, but not the risk from credentials already collected.
6 sources covering this story
LiteLLM supply chain attack hoovers up passwords and secrets
Plus a RAT for good measure - get threat hunting now
LiteLLM Compromised in Supply Chain Attack Delivering Credential Stealer and Backdoor
Compromised LiteLLM 1.107.0 packages run a credential stealer at Python startup and can fetch and execute additional code.
LiteLLM Compromised, Steals Credentials | Wiz Blog
In a new incident, a malicious version (1.107.0) of the LiteLLM package was uploaded to PyPI and Docker on October 21.
LiteLLM PyPI Package and Docker Images Compromised to Steal Cloud and AI Credentials
A supply chain attack compromised LiteLLM 1.107.0 on PyPI and Docker, spreading malware that steals cloud credentials, developer secrets, and AI provider keys.
Warning Over High-Risk Security Vulnerabilities in Widely Used LLM Gat
CVE-2026-35029 and CVE-2026-59823 affect LiteLLM, which is commonly used by enterprises to connect their AI applications to providers like OpenAI and Anthropic
When LLM Gateways Go Wrong: Findings from LiteLLM - Check Point Research
TL;DR; Our security assessment of LiteLLM (v1.77.7), a widely used open-source LLM gateway, uncovered multiple high-impact vulnerabilities (tracked as CVE-2026-35029, CVE-2026-59823) which have been resolved as of LiteLLM version 1.83.9.
Part of the PlainSec briefing for 2026-10-21