Vulnerability  ·  2026-06-23

LiteLLM Completions Interface — Incorrect Authorization via Banned Keywords Bypass (CVE-2026-12797)

VulnerabilityMedium impactGlobalCVE-2026-12797
A security flaw in BerriAI LiteLLM up to version 1.82.5 allows incorrect authorization in the async_pre_call_hook function of enterprise/enterprise_hooks/banned_keywords.py (Completions Interface). Manipulation of the prompt argument results in incorrect authorization, allowing bypassing of enterprise keyword-banning controls. CVSS 6.3 Medium; published 2026-06-21.
Enterprise deployments of LiteLLM use the banned-keywords hook as a security policy control to prevent specific prompt content from reaching LLM providers. A bypass defeats content-policy enforcement in AI gateways, enabling prompt injection of prohibited content or regulatory compliance violations in enterprise AI pipelines.
Crafted prompt argument sent to the Completions Interface bypasses the async_pre_call_hook keyword-ban enforcement.
LiteLLM (BerriAI) ≤ 1.82.5 (Enterprise edition, banned_keywords hook)
Upgrade LiteLLM to ≥ 1.84.0. NVD: https://nvd.nist.gov/vuln/detail/CVE-2026-12797
Sources
NVD CVE-2026-12797PoC Gist (YLChen-007)
See this in the live feed Explore related AI security and governance findings — updated every morning.
Open the feed →