All vulnerabilities

Servers and applications

GHSA-g5ff-637f-6q2m

internal_user_viewer arbitrary local file read via `vertex_ai_credentials`

Severity
High 8.1
Published
1 Oct 2026
Last changed
1 Oct 2026
Added to UpHeron
3 Nov 2026
Fixed in
1.95.0
Weakness
Improper input validation (CWE-20)
Attack
Over the network · Needs a user account · No user action needed
Attacker can
Read data · Change data
CVSS vector
CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:N
Sources
Project advisories (GitHub)

Description

An authenticated LiteLLM proxy user could read arbitrary local files from the proxy runtime by sending a crafted vertex_ai_credentials value to /utils/transform_request. The request-body safety blocklist rejected vertex_credentials but missed its alias vertex_ai_credentials, which let the value reach the Vertex AI credential loader and trigger a file read that is sent to an attacker-controlled URL.

A low-privilege authenticated user (only internal_user_viewer, obtainable through SSO login on many deployments) could read any file readable by the LiteLLM process. Reading files such as /proc/self/environ can disclose LITELLM_MASTER_KEY and provider API keys, so the impact can escalate to full proxy-admin control.

From BerriAI/litellm's security advisory.

Affected software

SoftwareAffected versions
LiteLLMServers and applications · github:berriai/litellm
  • Every version before 1.95.0

About LiteLLM

The fastest, litest AI Gateway. Rust core with Python SDK. Call 100+ LLM APIs in OpenAI (or native) format with cost tracking, guardrails, load balancing, and logging [Bedrock, Azure, OpenAI, Anthropic, OpenAI, VertexAI, vLLM, Nvidia NIM]

From its GitHub repository.
Weekly downloads (PyPI)
23,483,590PyPI Stats, 15 Nov 2026
GitHub stars
63,271GitHub, 15 Nov 2026
Forks
13,287GitHub, 15 Nov 2026
Latest release
v1.109.1GitHub, 15 Nov 2026
Released
12 Nov 2026GitHub
First published on GitHub
27 Jul 2023GitHub

References