Connect your agent
AdvisoriesCriticalCheck your lockfilePluginHow it worksPricing Connect your agent

Advisory record · GHSA-72m8-9m7m-h278

LiteLLM: Custom Code Guardrails production endpoints bypass code safety checks

CVSS 2.1 LowEPSS 0.0090 · 60th percentile

Published by the GitHub Advisory Database at 22 Jul 2026, 22:38 UTC. agentprime has not analysed this yet, so whether it affects any app is not known.

PackageEcosystemAffectedFixed in
litellmPyPI<1.82.01.82.0
Identifiers
GHSA-72m8-9m7m-h278 · CVE-2026-59821 · PYSEC-2026-3478
Weakness (CWE)
CWE-94
Tags
Published
22 Jul 2026
Updated
10 Sept 2026

Machine-written summary

Summary written by Claude Haiku from the source advisory. Not reviewed by a person.

LiteLLM versions before 1.82.0 have a flaw where the production create and update paths for Custom Code Guardrails skip the sandboxing and validation applied by the test endpoint. An attacker who holds a privileged account able to create or update guardrails can run custom Python code inside the LiteLLM proxy, which may expose secrets available to the process. Upgrade to LiteLLM 1.82.0 or later. Until then, limit access to the guardrail create and update endpoints to trusted administrators, make sure a master key is set, and avoid enabling Custom Code Guardrails for untrusted users.

Written 1 Nov 2026.

Not analysed yet

We haven’t published our analysis of this Advisory yet. Once a person has reviewed it, this page will say what’s vulnerable, what has to be true for your app to be affected, and how to check.

Severity and exploitation

ScoreRatingCVSSVectorSource
2.1 Low 4.0 CVSS:4.0/AV:N/AC:L/AT:P/PR:H/UI:N/VC:L/VI:L/VA:L/SC:N/SI:N/SA:N CNA (GitHub_M)
  • CISA KEV Not listed.
  • EPSS 0.0090, the 60th percentile, on 13 Nov 2026: FIRST’s estimate of the chance of exploitation in the next 30 days.

Timeline

  1. 31 Oct 2026, 14:11 UTC First seen by agentprime
  2. 31 Oct 2026, 14:11 UTC Confirmed by OSV

Sources

Facts on this page come from the GitHub Advisory Database (CC-BY 4.0), OSV and the other sources listed, as published. We don’t reproduce full advisory descriptions, which can contain exploit code.