AI agent vulnerability · LiteLLM

CVE-2026-35030: LiteLLM: Authentication bypass via OIDC userinfo cache key collision

criticalCVSS 4.0 9.4GHSA-jjhc-v7c2-5hh6

Impact When JWT authentication is enabled (enablejwtauth: true), the OIDC userinfo cache uses token[:20] as the cache key. JWT headers produced by the same signing algorithm generate identical first 20 characters. This configuration option is not enabled by default. Most instances are not affected. An unauthenticated attacker can craft a token whose first 20 characters match a legitimate user's cached token.

Published Apr 3, 2026 · Updated Apr 6, 2026

Affected packages

litellm

Ecosystem: PyPI
Affected: < 1.83.0
Fixed in: 1.83.0
Check: pip show litellm

Advisory details

From the GitHub Advisory Database (CC BY 4.0).

Impact

When JWT authentication is enabled (enable_jwt_auth: true), the OIDC userinfo cache uses token[:20] as the cache key. JWT headers produced by the same signing algorithm generate identical first 20 characters.

This configuration option is not enabled by default. Most instances are not affected.

An unauthenticated attacker can craft a token whose first 20 characters match a legitimate user's cached token. On cache hit, the attacker inherits the legitimate user's identity and permissions. This affects deployments with JWT/OIDC authentication enabled.

Patches

Fixed in v1.83.0. The cache key now uses the full hash of the JWT token.

Workarounds

Disable OIDC userinfo caching by setting the cache TTL to 0, or disable JWT authentication entirely.

Credited to veria-labs.

References

How this page is made: the data panel comes from the GitHub Advisory Database, CISA’s KEV catalog, the NVD and FIRST EPSS, and updates twice a day. No analysis has been written for this entry yet. Methodology

More LiteLLM vulnerabilities

All LiteLLM entries