Cyber Resilience

CVE-2026-35030

Auth Bypass in Litellm ≤ 1.83.0

Published
06 April 2026
Modified
15 July 2026
Patch / advisory
CVSS Score v4 9.4
Click a component to see what it means
Raw vectorCVSS: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/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X
EPSS Score 0.0049 40th percentile
Risk Priority 46 floored blend · peak EPSS

Summary

CVE-2026-35030 is a critical-severity Improper Authentication (CWE-287) vulnerability in Litellm Litellm. Its CVSS base score is 9.4 (Critical).

Operationally, exploitation aligns with the MITRE ATT&CK technique External Remote Services (T1133); ranked at the 40th percentile by exploit likelihood (below the median); it is not currently listed in the CISA KEV catalog.

This vulnerability is AI-related — categorised as LLM Application Platforms; in the Other ATLAS/OWASP Terms risk domain.

The strongest mitigations our analysis identified map to AU-4 (Audit Log Storage Capacity) and IA-2 (Identification and Authentication (Organizational Users)) — see the control section below for these in your framework.

Deeper analysis AI-assisted summary

Synthesised by an AI model from the NVD description and linked references — a reading aid, not an authoritative source.

CVE-2026-35030 is a high-severity authentication bypass vulnerability (CVSS 9.1, CWE-287) in LiteLLM, an open-source proxy server and AI gateway for calling LLM APIs in OpenAI or native formats. The issue affects versions prior to 1.83.0 when JWT authentication is explicitly enabled via the `enable_jwt_auth: true` configuration option, which is not enabled by default. In these configurations, the OIDC userinfo cache improperly uses the first 20 characters of the JWT token (`token[:20]`) as the cache key. JWT headers generated by the same signing algorithm produce identical first 20 characters, enabling predictable cache key collisions.

An unauthenticated network attacker can exploit this vulnerability with low complexity and no privileges required by crafting a malicious JWT token whose first 20 characters match those of a legitimate user's cached token. Upon a cache hit, the attacker inherits the legitimate user's identity and associated permissions, potentially gaining unauthorized high-confidentiality and high-integrity access to the LiteLLM proxy and downstream LLM APIs.

The vulnerability is addressed in LiteLLM version 1.83.0, which fixes the cache key generation to prevent collisions. Security practitioners should upgrade to v1.83.0 or later and review configurations to confirm JWT/OIDC authentication usage, as most deployments remain unaffected due to the non-default setting. Additional details are available in the GitHub Security Advisory at https://github.com/BerriAI/litellm/security/advisories/GHSA-jjhc-v7c2-5hh6.

OWASP Top 10 for Web (2025)

EU & UK References

Vulnerability Data

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…

more

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.

CWE(s)

AI Security AnalysisAI

AI Category
LLM Application Platforms
Risk Domain
Other ATLAS/OWASP Terms
OWASP Top 10 for LLMs 2025
None mapped
Classification Reason
Matched keywords: ai, llm, openai

Related Threats

MITRE ATT&CK Enterprise Techniques

T1133 External Remote Services Persistence
Adversaries may leverage external-facing remote services to initially access and/or persist within a network.
T1190 Exploit Public-Facing Application Initial Access
Adversaries may attempt to exploit a weakness in an Internet-facing host or system to initially access a network.
T1040 Network Sniffing Credential Access
Adversaries may passively sniff network traffic to capture information about an environment, including authentication material passed over the network.
T1134 Access Token Manipulation Stealth
Adversaries may modify access tokens to operate under a different user or system security context to perform actions and bypass access controls.
T1185 Browser Session Hijacking Collection
Adversaries may take advantage of security vulnerabilities and inherent functionality in browser software to change content, modify user-behaviors, and intercept information as part of various browser session hijacking techniques.
T1505.003 Web Shell Persistence
Adversaries may backdoor web servers with web shells to establish persistent access to systems.
Derived from this CVE’s CWE(s) via the direct CWE→ATT&CK cross-walk.

CVEs Like This One

CVE-2026-59822Same product: Litellm Litellm
CVE-2026-12773Same product: Litellm Litellm
CVE-2026-12795Same product: Litellm Litellm
CVE-2026-35029Same product: Litellm Litellm
CVE-2026-49468Same product: Litellm Litellm
CVE-2026-59819Same product: Litellm Litellm
CVE-2026-42203Same product: Litellm Litellm
CVE-2024-5710Same product: Litellm Litellm
CVE-2024-4888Same product: Litellm Litellm
CVE-2026-47101Same product: Litellm Litellm

Affected Assets

litellm
litellm
≤ 1.83.0

Mitigating Controls

Control response

Prevent
Stop it (NIST 800-53)

Detect
Catch it (NIST detect / respond)

Harden
Shrink the surface (DISA STIG)
  • 8 hardening rules · 5 OS baselines
Validate
Prove the fix (OWASP ASVS)
  • V6.4.4
  • V6.5.4
  • V6.5.5
  • V6.5.7

Mitigating Controls (NIST 800-53 r5) AI

Mandates unique identification and authentication of organizational users before access, directly stopping improper authentication.

Requires unique identification and authentication of devices before establishing connections, preventing unauthenticated device claims.

Mandates unique identification and authentication of non-organizational users, directly addressing the weakness for external actors.

Allocating sufficient audit log storage capacity directly stops truncation of security-relevant records due to overflow.

Manages authenticators with verification and secure distribution, reducing opportunities for improper authentication.

Enforces access only after approved authorizations, which presupposes correct authentication has occurred.

Mitigating Controls (NIST CSF 2.0) AI

Derived directly from the weakness types (CWEs) cited in the NVD entry via our AI-authored CWE→CSF cross-walk (authority under review) — links open the control.

PR.AA-03 mostly match
prevents

PR.AA-03 directly enforces authentication mechanisms that eliminate most improper-authentication defects, yet CWE-287 spans additional vectors (missing checks, flawed protocols, session handling) that one control does not fully close.

PR.AA-04 mostly match
prevents

PR.AA-04 directly enables verification of identity assertions (mostly preventing CWE-287 in that scope) yet leaves many other authentication failure modes unaddressed (only partial prevention overall).

PR.PS-04 mostly match
prevents

Requiring complete log generation directly prevents truncation of security-relevant records.

DE.AE-03 partial match
prevents

Correlating logs from multiple sources can surface truncation only if the logs themselves are complete.

DE.CM-09 partial match
prevents

Runtime and data monitoring depends on receiving full security-relevant information without truncation.

PR.AA-01 partial match
prevents

PR.AA-01 supplies and governs credentials/tokens that authentication relies on, removing some weak-credential cases, yet leaves verification logic, missing checks, and protocol flaws untouched.

Mitigating Controls (ISO/IEC 27001:2022 Annex A) AI

Derived directly from the weakness types (CWEs) cited in the NVD entry via our AI-authored CWE→ISO cross-walk (authority under review) — links open the control.

prevents

Requiring authentication methods whose strength matches the sensitivity of the data and mandating multi-factor authentication directly blocks attempts to access resources without proving identity.

prevents

Verifying user identity before issuing replacement credentials and forcing changes after compromise reduces the likelihood that authentication bypass can be achieved through stolen or weak credentials.

prevents

Mandating segregated approval and oversight for non-human identities reduces the chance that weak or orphaned credentials can be exploited for unauthorized authentication.

A.8.15 Logging partial match
prevents

Logging ensures security-relevant events are recorded without truncation that could hide attack details.

degrades

Monitoring activities rely on complete event data; truncation would impair detection and analysis.

mitigates

Requiring authentication mechanisms and technical parameters for secure connections ensures that network services verify user identity before granting access, preventing exploitation of missing or weak authentication.

Hardening callouts derived

Configuration rules from DISA STIG baselines that bear on weaknesses of the type cited by this CVE. Each rule is shown with the relationship its mapping actually records, against the CWE it was authored against. Derived via CVE→CWE over `controls_xwalks` (authoritative rows only; rows rated `none` are excluded).

Oracle Linux 8 (1 rule)
  • V-248827 OL 8 must not have the rsh-server package installed. prevents CWE-287
RHEL 7 (3 rules)
  • V-204425 The Red Hat Enterprise Linux operating system must be configured so that the SSH daemon does not allow authentication using an empty password. prevents CWE-287
  • V-204442 The Red Hat Enterprise Linux operating system must not have the rsh-server package installed. prevents CWE-287
  • V-204424 The Red Hat Enterprise Linux operating system must not allow accounts configured with blank or null passwords. prevents CWE-287
RHEL 8 (1 rule)
  • V-230492 RHEL 8 must not have the rsh-server package installed. prevents CWE-287
Ubuntu 22.04 (1 rule)
  • V-260470 Ubuntu 22.04 LTS, when booted, must require authentication upon booting into single-user and maintenance modes. prevents CWE-287
Ubuntu 24.04 (2 rules)
  • V-270675 Ubuntu 24.04 LTS when booted must require authentication upon booting into single-user and maintenance modes. prevents CWE-287
  • V-270714 Ubuntu 24.04 LTS must not allow accounts configured in Pluggable Authentication Modules (PAM) with blank or null passwords. prevents CWE-287

References