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

Should you act?

Critical and remotely reachable

No known exploitation yet, but it is network-reachable without credentials. This signal is computed from CISA KEV, EPSS and CVSS data; it is not an editorial verdict.

• No confirmed exploitation. It is not in CISA’s KEV catalog and CISA’s SSVC assessment records no known exploitation. That can change quickly after a public write-up; re-check if you are affected.
• EPSS 0.9%, percentile 59: FIRST’s model estimates a 0.9% chance of exploitation activity in the next 30 days, higher than 59% of all scored CVEs (moderate). EPSS is a probability, not a verdict.
• Remotely reachable with no login and no user interaction. If the vulnerable service is exposed to the internet, prioritize it.
• Only applies if your dependency tree includes litellm (PyPI) < 1.83.0, fixed in 1.83.0. Check your lockfile, including transitive dependencies.

What it is

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 (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. 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.

From the GitHub advisory

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. On cache hit, the attacker inherits the legitimate user's identity and permissions. This affects deployments with JWT/OIDC authentication enabled.

Patches

Weakness type: CWE-287, CWE-222.

Product: litellm (PyPI) · Assigned by: GitHub_M · Published: 2026-04-06

Affected and fixed versions

Severity scores

• Attack vector: network (remotely reachable)
• Attack complexity: low
• Attack requirements: present (depends on deployment conditions)
• Privileges required: none (no login)
• User interaction: none

CVSS measures how bad a flaw is if exploited, not how likely exploitation is. Read it next to the exploitation data below.

Exploitation and urgency

• CISA KEV: not listed (no confirmed in-the-wild exploitation known to CISA).
• CISA SSVC assessment: exploitation none, automatable yes, technical impact total.
• FIRST EPSS (model of 2026-10-21): 0.88% probability of exploitation activity in the next 30 days; percentile 59.

Timeline

• 2026-04-03 — GitHub-reviewed advisory GHSA-jjhc-v7c2-5hh6 published (critical)
• 2026-04-06 — CVE record published by GitHub_M
• 2026-07-15 — CVE record last updated

Sources and references

Original sources first. Version and fix details should always be confirmed against the vendor or project advisory.

Vendor and project advisories, patches

1. access.redhat.com access.redhat.com
2. access.redhat.com access.redhat.com
3. access.redhat.com access.redhat.com

Vulnerability records

1. GitHub Advisory GHSA-jjhc-v7c2-5hh6 github.com
2. CVE.org record CVE-2026-35030 cve.org
3. NVD entry CVE-2026-35030 nvd.nist.gov

Other references from the CVE record and advisory

1. github.com github.com
2. access.redhat.com access.redhat.com
3. RHBZ#2455509 bugzilla.redhat.com
4. security.access.redhat.com security.access.redhat.com

Critical PyPI advisories · Today’s Security Digest