GHSA-g5ff-637f-6q2mHigh· 8.1▾ TwilightLiteLLM: internal_user_viewer arbitrary local file read via `vertex_ai_credentials`
▾ Twilight zone — High severity, or a signal on a lesser flaw
impact 44.6 · likelihood 0 · exploitation 0
Need a working PoC? Pro members can cast a request and our team develops one — it lands right here.
An authenticated LiteLLM proxy user could read arbitrary local files from the proxy runtime by sending a crafted vertex_ai_credentials value to /utils/transform_request. The request-body safety blocklist rejected vertex_credentials but missed its alias vertex_ai_credentials, which let the value reach the Vertex AI credential loader and trigger a file read that is sent to an attacker-controlled URL.
A low-privilege authenticated user (only internal_user_viewer, obtainable through SSO login on many deployments) could read any file readable by the LiteLLM process. Reading files such as /proc/self/environ can disclose LITELLM_MASTER_KEY and provider API keys, so the impact can escalate to full proxy-admin control.
Upgrade to 1.95.0 or later. The fix adds the missing alias to the request-body blocklist, so no configuration change is needed once you upgrade. There is no reliable configuration workaround on affected versions.
litellm < 1.95.0Upgrade to a patched release:
litellm 1.95.0Connected by shared product, vendor, weakness, or advisory.
CVE-2026-59820Medium· 6.5LiteLLM is a proxy server (AI Gateway) to call LLM APIs in OpenAI (or native) format
GHSA-hhww-mrg2-969hMedium· 6.8LiteLLM vulnerable to reflected XSS in /sso/debug/callback
CVE-2026-114612Critical· 9.9LiteLLM: Privilege Escalation to Proxy Admin via Cross-Domain Reuse of the Salt Key
CVE-2026-59823Medium· 5.3LiteLLM is a proxy server (AI Gateway) to call LLM APIs in OpenAI (or native) format
CVE-2026-12799Medium· 4.3BerriAI litellm: UI User Enumeration leads to System-Wide Information Disclosure
CVE-2026-12796Medium· 6.3BerriAI litellm: SSO Login Does Not Invalidate Previous UI Session Tokens