CVE-2026-35030
An unauthenticated attacker can craft a token whose first 20 characters match a legitimate user's cached token.
Does this matter?
High impact if exploited, but EPSS currently rates exploitation as unlikely (0.49%). Schedule it in the normal patch cycle and watch for a rise in EPSS or a public exploit.
Description
LiteLLM is a proxy server (AI Gateway) to call LLM APIs in OpenAI (or native) format. Prior to 1.83.0, 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. Fixed in v1.83.0.
- CVSS 4.0
- 9.4 CRITICALCVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:N/VC:H/VI:H/VA:N/SC:H/SI:H/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/M
- EPSS
- 0.49% probability · 41th percentile
- CISA KEV
- Not listed
- Weakness
- CWE-287, CWE-222
- Affected
- litellm/litellm
- Source
- security-advisories@github.com
References
- https://github.com/BerriAI/litellm/security/advisories/GHSA-jjhc-v7c2-5hh6Mitigation, Vendor Advisory
- https://access.redhat.com/errata/RHSA-2026:13545
- https://access.redhat.com/errata/RHSA-2026:28960
- https://access.redhat.com/errata/RHSA-2026:30056
- https://access.redhat.com/security/cve/CVE-2026-35030
- https://bugzilla.redhat.com/show_bug.cgi?id=2455509
- https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-35030.json
Source: NVD record, EPSS from FIRST.org, KEV from CISA. Refreshed daily.