LiteLLM < 1.84.0 Improper Authentication Vulnerability (GHSA-7488-6r32-c95q)

The version of the LiteLLM Python package installed on the remote host is prior to 1.84.0. It is, therefore, affected by a vulnerability: - LiteLLM contains an improper authentication vulnerability in the MCP Streamable HTTP endpoint. When LiteLLM key validation fails, the fallback OAuth2...

CVE-2026-42208-Lab

CVE-2026-42208 — LiteLLM Pre-Auth SQL Injection Timing PoC Local Docker lab and least-harm PoC for CVE-2026-42208, a pre-authentication SQL injection vulnerability in LiteLLM Proxy's API key verification path. This repository demonstrates the difference between a vulnerable LiteLLM instance and a...

CVE-2026-42208_lab

LiteLLM Proxy SQL Injection GHSA-r75f-5x8p-qvmc A reproduction environment for the SQL injection vulnerability in LiteLLM Proxy's API key authentication flow. Vulnerability Summary Item| Detail ---|--- Advisory| GHSA-r75f-5x8p-qvmc Type| SQL Injection CWE-89 Severity| Critical Affected| litellm...

threat-intel-brief-cve-2026-42208-litellm

Threat Intelligence Brief - CVE-2026-42208: BerriAI LiteLLM SQL Injection CVE: CVE-2026-42208 GHSA: GHSA-r75f-5x8p-qvmc CVSS Score: 9.3 Critical Affected Software: BerriAI LiteLLM versions = 1.81.16, 1.83.7 Fixed Version: 1.83.7-stable released April 19, 2026 CISA KEV Added: May 08, 2026 Sources:...

CVE-2026-42208

✦ CVE-2026-42208 - Critical SQL Injection in LiteLLM 🧭 Overview CVE-2026-42208 is a critical pre-authentication SQL injection vulnerability affecting the LiteLLM AI gateway. It allows attackers to exploit backend database queries without authentication , potentially exposing sensitive AI...

poc_cve-2026-42208

CVE-2026-42208 — LiteLLM Pre-Authentication SQL Injection A lab environment for reproducing and detecting CVE-2026-42208 , a critical pre-authentication SQL injection vulnerability in LiteLLM where unsanitized Bearer tokens reach a raw PostgreSQL query. Vulnerability Overview Field| Details ---|-...

PoeLLM Malware Infects 3,400+ Servers to Expand Crypto Mining Botnet

Cybersecurity researchers are calling attention to a new malware family that has been observed targeting exposed artificial intelligence AI and large language model LLM infrastructure with an aim to deploy cryptocurrency miners and further expand the scale of the botnet. The financially motivated...

Exploit for CVE-2026-93355

CVE-2026-93355 — LiteLLM unverified-email JWT account takeover...

LiteLLM - Command Injection

A critical unauthenticated remote code execution vulnerability exists in LiteLLM due to improper input handling in the MCP stdio test endpoint. An attacker can send a specially crafted request to the /mcp-rest/test/connection endpoint with controlled parameters, resulting in arbitrary command...

CVE-2026-40217-PoC

CVE-2026-40217 — LiteLLM Guardrail Sandbox Escape LiteLLM POST /guardrails/testcustomcode — Sandbox Escape leading to Remote Code Execution RCE as root in default Docker deployment. Field| Value ---|--- CVE| CVE-2026-40217 CVSS| 8.8 HIGH — CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H CWE| CWE-913...

CVE-LiteLLM

cve-id ⚡ Simple Usage Use this project only in safe and authorized environments such as: Local virtual machines Docker containers Isolated lab setups Authorized penetration testing environments Example setup: git clone cd Project use python follow below Create a virtual environment python -m venv...

CVE-2026-47101-PoC

CVE-2026-47101 — LiteLLM Privilege Escalation via /key/generate + /user/update LiteLLM v1.82.6 before v1.83.14 /key/generate endpoint allows low-privileged internaluser to request an API key with wildcard routes "/", and then elevate their own role to proxyadmin via the /user/update endpoint,...

CVE-2025-45809-PoC

CVE-2025-45809 — LiteLLM SQL Injection via /key/block Time-Based Blind SQLi LiteLLM v1.65.4 versions before v1.81.0 /key/block and /key/unblock endpoints have a SQL injection vulnerability in the key parameter. Attackers can use time-based blind injection techniques to steal database content and...

CVE-2025-11203-PoC

CVE-2025-11203 — LiteLLM Health Endpoint APIKEY Information Disclosure LiteLLM versions 1.63.14 /health endpoint, when processing the apikey parameter, does not properly filter sensitive information , causing authenticated users to retrieve API Keys stored in other model configurations. The apike...

CVE-2026-47102-PoC

CVE-2026-47102 — LiteLLM Privilege Escalation via /user/update LiteLLM v1.83.7 versions before v1.83.10 /user/update endpoint allows low-privileged users with access to this endpoint to modify their own userrole field to proxyadmin when updating their account, achieving unauthorized privilege...

CVE-2026-35030-PoC

CVE-2026-35030 — LiteLLM Authentication Bypass via OIDC Userinfo Cache Key Collision LiteLLM OIDC userinfo cache uses token:20 as the cache key. Two different JWTs signed with the same algorithm produce identical first 20 characters , allowing an unauthenticated attacker to inherit another user's...

CVE-2026-35029-PoC

CVE-2026-35029 — LiteLLM /config/update Broken Access Control LiteLLM 's /config/update endpoint does not check the caller's role privileges. Any authenticated user with a valid API Key without admin privileges can modify the proxy configuration, register malicious Pass-Through endpoints to achie...

litellm-sqli-scanner

LiteLLM SQL Injection Scanner CVE-2026-42208 — SQL injection in LiteLLM Proxy Detects and validates CVE-2026-42208: SQL injection vulnerability in LiteLLM Proxy's database layer. For authorized security testing only. Features CVE-2026-42208 SQL injection detection LiteLLM Proxy version...

CVE-2026-49468-LiteLLM-Auth-Bypass

CVE-2026-49468 — LiteLLM Unauthenticated Auth Bypass via Host-Header Route Confusion Pre-authentication authentication/authorization bypass in the LiteLLM proxy BerriAI. A single crafted Host header makes the proxy evaluate its auth decision against a public health route while FastAPI still...

Cooldown Landmines: Cross-Tenant Interference Attacks on LLM Gateways

LLM gateways enforce separate tenant quotas while sharing model deployments and cooldown records that temporarily exclude failing backends. However, a tenant's request failure can update these shared records and restrict other tenants' access to serviceable deployments. We identify two attacks th...