← 返回列表

LiteLLM Compromised, Steals Credentials

单站 Wiz Blog 作者 Benjamin Read 发布 Wed, 21 Oct 2026 13:44:12 GMT 入库 2026-11-18T05:29:57

At 11:39 UTC on October 21, a malicious version (1.107.0) of the LiteLLM package was uploaded to PyPI. This version was quarantined on PyPI as of 13:20 UTC. According to ClickPy data, the malicious version received approximately 49,000 downloads.

LiteLLM Docker containers on GHCR and Docker hub also included the trojanized version of LiteLLM. These included the v1.107.0 , latest, main-latest and main-stable tags. These were removed \~13:53 UTC.

The most recent safe version is v1.106.2.

When the malicious package is installed, three files are extracted directly into the site-packages root: cook.pth, cook.py and sitecustomize.py. These packages are designed to be executed whenever a python interpreter is started.

Per LiteLLM developer BerriAI:

malicious actions were taken through three engineers' GitHub accounts, appearing to be by way of stolen GitHub CLI OAuth tokens. The threat actor used one user's account triggered the project-releaser run, and another user's account to approve it. All tokens which were involved in this attack have been revoked.

And in a followup:

GitHub CLI OAuth token exposure After a thorough investigation, we have observed specific situations in which Claude Code will embed a user’s GitHub CLI OAuth token into a response body, and in situations where response bodies are logged, these tokens may appear in LiteLLM logs. Access to our internal LiteLLM deployment We identified an unexpected, innocuously named user account in our internal LiteLLM deployment, which was exposed to the public internet. We believe the attacker gained access to this deployment, obtained the GitHub CLI OAuth tokens from its logs, and subsequently used those credentials to compromise our release workflow. We have not yet confirmed how the attacker gained access to our internal deployment, but currently suspect it was through a recently patched vulnerability.

UPDATES

The payload has strong indications of LLM-aided development, including verbose variable names (encrypted_private_key_output), overly defensive error handling with generic try/except blocks, platform-specific code paths that fail silently on untested platforms (Windows marker path /tmp/ doesn't exist, fcntl/fork calls crash the remote loader), and a dead code module (exec_py) that was defined but never called. The structure suggests iterative prompting: functional on the happy path but brittle at the edges and untested.

Upon startup, the interpreter processes cook.pth, which includes a command to import cook, which loads cook.py. cook.py deobfuscates itself, runs a stealer routine and writes a maker to /tmp/.pc_once_72129118c59f5636c0bb (note this fails on Windows, as the /tmp/ path often does not exist). The stealer routine gathers system, browser, developer, cloud and AI tool secrets, encrypts them and exfiltrates to www.litellm.dev/healthcheck via a post request. This domain is attacker controlled, impersonating LiteLLM infrastructure.

Python interpreters will also attempt to execute sitecustomize.py - however, it will only run on Linux & MacOS machines. This script attempts to fetch domain.txt from a GitHub dead drop (letters9813/pythonclient-test/main/domain.txt). If it receives a response, it will attempt to download a second stage python script from http://<domain>/codepy and run it. As of the time of analysis, the domain.txt file was never populated beyond initial test values. The repo has been taken down by GitHub as of 1:30 PM UTC.

The dual-technique approach, using both .pth files and sitecustomize.py simultaneously, appears to be the first documented case in the wild.

The embedded credential stealer, internally named privcollect, harvests sensitive data across 16 modules targeting browsers, cloud providers, and AI services. It extracts:

In terms of AI-specific theft, it targets API keys for OpenAI, Anthropic, Cohere, and other LLM providers stored in environment variables or config files. It also collects SSH private keys, GPG keys, and shell history files.

LiteLLM Proxy stores upstream API keys in a PostgreSQL database. The stealer queries this database directly (SELECT * FROM "LiteLLM_CredentialsTable" LIMIT 500) to extract all configured provider credentials.

Harvested data is encrypted with AES-256-CBC (key wrapped with RSA-4096) and exfiltrated over HTTP to www.litellm[.]dev/healthcheck using the hardcoded X-Token header and an Internet Explorer 7 user-agent string.

The payload originates with malicious commits made on github.com/Berri-AI/litellm. Note, this is a hyphenated variant of the legitimate BerriAI organization. This repository, not visible on GitHub, was used to inject the payload into the release pipeline.

Between 08:56 and 10:46 UTC on October 21, the attacker pushed 10 commits using the author email ckdi82fkoos@proton.me. The first nine commits were created on GitHub.com and therefore signed with GitHub's verified signature.

The first three (within 2 minutes) added the persistence files: cook.pth, cook.py, and sitecustomize.py. Four more commits bumped version numbers to 1.107.0.

After a 39-minute gap, two commits refined the payload and added the critical data = "cook" directive to pyproject.toml, which instructed the build tool to bundle the malware into the wheel.

A final commit, made 58 minutes later with a different email (Berri-AI@users.noreply.github.com) and unsigned (suggesting a different environment), added COPY cook ./cook to the Dockerfiles.

Although the Berri-AI repository is unavailable, all commits remain accessible through the parent repository due to Git's fork object sharing. The full diff is available at: github.com/BerriAI/litellm/compare/a8113b0d...20535d8f

Wiz Research has observed abuse of stolen credentials, starting approximately 4.5 hours post-exfiltration. Between 16:20 and 17:01 UTC, the attacker validated AWS instance role credentials harvested from compromised CI runners and Kubernetes nodes across multiple environments, from the following IP addresses:

Each credential was checked with sts:GetCallerIdentity, followed by bedrock:ListFoundationModels and a ListModels call, consistent with LLMjacking.

In a small number of environments, the attacker also enumerated IAM (ListRoles, ListUsers, ListAccessKeys) and performed general discovery (ListAccounts, DescribeTrails, etc.).

Wiz customers should review our Threat Intel Center advisory: Supply Chain Attack Targeting LiteLLM .

Indicator Description Type
litellm[.]dev Attacker-controlled exfil (lookalike) Network
www.litellm[.]dev/healthcheck Exfiltration endpoint Network
http://{domain}/codepy Dynamic second-stage RCE payload (never observed active) Network
X-Token: XWiA8gZNa7NcX2XRoAEb Hardcoded exfil auth header Network
/tmp/.pc_once_72129118c59f5636c0bb Run-once marker file Host
PRIVCOLLECT_ACTIVE=1 Environment variable indicator Host
84dc8c82c0a4a4a62552ef826b49fcc86d1a5c4a5d0f2c06665723f93d129d7d cook.pth persistence hook - Windows Hash
7f191632e7320705d154b85204f1ae546b543acbd7e8861229b76f5042b92bdb cook.pth persistence hook - Linux/MacOS Hash
03ac2a8b752605d1d7710979c17ff18b03dd978203ba23ba97fd8ce3d6f897ea cook.py credential stealer payload (22,252 bytes) - Windows Hash
9dd0a3926f96a37eb86cf1ee2310d9be93139cc4ebef3ef9e2efe9531390be25 cook.py credential stealer payload (22,252 bytes) - Linux/MacOS Hash
b6122440d0d9cb1eb1ade9933f49214da8c31288de7710c076c2680dfa528465 sitecustomize.py loader with hourly backdoor loop (25,509 bytes) - Windows Hash
34bef04fe88ac9d0aae0475630b1738681d15b94f233ac8aa46eacf5968305fa sitecustomize.py loader with hourly backdoor loop (25,509 bytes) - Linux/MacOS Hash

File/Image SHA256
litellm-1.107.0-cp310-abi3-macosx_10_12_x86_64.whl 7598727f209d717041db37f7dca271d3a7d328a279890e0e1d55375729aeb86d
litellm-1.107.0-cp310-abi3-macosx_11_0_arm64.whl 0aaf7ba346d41ef2f40390436738b001180d5a1c03aa093f121078a9969d9a1a
litellm-1.107.0-cp310-abi3-manylinux_2_28_aarch64.whl 668169d35b87bf51324080390b1b1c5ab12bd335cfe5f1a99e3689abd5e02ae6
litellm-1.107.0-cp310-abi3-manylinux_2_28_x86_64.whl 462d0c9f67ca522d9e0e4acd245e90b36ce0d5353604a66bb7c43a1ec4a6fcf7
litellm-1.107.0-cp310-abi3-musllinux_1_2_aarch64.whl c2e32d0ce2548aa662c18bf5001911a79fe5b30cf8357132275add02fdcde054
litellm-1.107.0-cp310-abi3-musllinux_1_2_x86_64.whl 1bbce8caa94f01c0aaa0b7ddae51553a8a32e1ce53d96d44fe583272aff7ad00
litellm-1.107.0-cp310-abi3-win_amd64.whl 12d4ccc5c6f8d35cf4f2cf233b7f3e418e863e657d81ff5f5fc9e210005b9484
ghcr.io/berriai/litellm:v1.107.0 linux/amd64 bd1aeee419ea36da9aa2b657288da4101887be468f020ac63d22ba6ed84a3ea5
ghcr.io/berriai/litellm:v1.107.0 linux/arm64 56b74573eed6e4aafe4b50603dbe7853e0345f188c8e44a09f1098caf1f4bcdf
ghcr.io/berriai/litellm-database:v1.107.0 linux/amd64 24581ddc30767f16ecb69d2b0517556f2cd7dea56971adb42a3d1beb97d69523
ghcr.io/berriai/litellm-database:v1.107.0 linux/arm64 827dd893c74cf621a860a66a3a447a5c732533f087589ce905573e285c0a239c
ghcr.io/berriai/litellm-non_root:v1.107.0 linux/amd64 645a67d810628252d63df85dcc8824c58bcc77cf8383f4d3e6321f0ccc3ce9e4
ghcr.io/berriai/litellm-non_root:v1.107.0 linux/arm64 f75c8edae59b72b1ebfd617295833f465c85bb2fab9c1d60f21b496db86cb8da
docker.io/litellm/litellm linux/amd64 bd1aeee419ea36da9aa2b657288da4101887be468f020ac63d22ba6ed84a3ea5
docker.io/litellm/litellm linux/arm64 56b74573eed6e4aafe4b50603dbe7853e0345f188c8e44a09f1098caf1f4bcdf
原文链接:https://www.wiz.io/blog/litellm-compromised-steals-credentials