← Vulnerability feed

Vulnerability record · CVE-2026-12797 · published 21 June 2026

CVE-2026-12797: Litellm improper authorization vulnerability

Litellm · Litellm

A security flaw has been discovered in BerriAI litellm up to 1.82.5. Affected is the function async_pre_call_hook of the file enterprise/enterprise_hooks/banned_keywords.py of the component Completions Interface. The manipulation of the argument prompt results in incorrect authorization. The attack may be performed from remote. The exploit has been released to the public and may be used for attacks. The vendor was contacted early about this disclosure.

2.1 CVSS 4.0 Low EPSS 0.40% · top 67.9% CWE-285 · Improper authorizationCWE-863 · Incorrect authorization
2.1CVSS 4.0 base score, v2 6.5
0.40%EPSS exploitation probability, 30 days
NoNot in CISA KEV
1Affected product versions listed by NVD
5References, 2 tagged exploit
24 Jun 2026Last modified by NVD

Description

A security flaw has been discovered in BerriAI litellm up to 1.82.5. Affected is the function async_pre_call_hook of the file enterprise/enterprise_hooks/banned_keywords.py of the component Completions Interface. The manipulation of the argument prompt results in incorrect authorization. The attack may be performed from remote. The exploit has been released to the public and may be used for attacks. The vendor was contacted early about this disclosure.

CVSS:4.0/AV:N/AC:L/AT:N/PR:L/UI:N/VC:L/VI:L/VA:L/SC:N/SI:N/SA:N/E:P/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:

Affected products

1 vulnerable configurations from NVD's CPE data, grouped by vendor and product.

References

LinkTags
https://gist.github.com/YLChen-007/078179224f07cc4e39e4f141a18c817a ExploitThird Party Advisory
https://vuldb.com/cve/CVE-2026-12797 Third Party AdvisoryVDB Entry
https://vuldb.com/submit/811288 Third Party AdvisoryExploitVDB Entry
https://vuldb.com/vuln/372559 Third Party AdvisoryVDB Entry
https://vuldb.com/vuln/372559/cti Permissions RequiredVDB Entry

Track CVE-2026-12797 inside VULONE

Watch it alongside the ransomware crews, C2 infrastructure and forum chatter that reference it, query it through the API and pull it into your SIEM over TAXII.

Start free Open in platform

Related vulnerabilities

Same products first, then exploited flaws of the same weakness class.

9.4CVE-2026-33634Trivy and trivy-action supply chain compromise via malicious release and tagsA threat actor used compromised credentials to publish a malicious Trivy v0.69.4 release, force-push 76 of 77 version tags in aquasecurity/trivy-acti…KEVEPSS 1.7%analysed9.3CVE-2026-42208LiteLLM proxy SQL injection in API key checkLiteLLM versions 1.81.16 to before 1.83.7 build a database query for proxy API key checks by concatenating the caller-supplied key into the query tex…KEVEPSS 5.8%analysed8.8CVE-2026-59822LiteLLM MCP endpoint auth bypass via OAuth2 passthrough fallbackLiteLLM's MCP Streamable HTTP endpoint, prior to 1.84.0, let an unauthenticated attacker send a fabricated Authorization header that triggered an OAu…KEVEPSS 0.84%analysed8.7CVE-2026-42271LiteLLM MCP test endpoints allow authenticated OS command injectionLiteLLM versions 1.74.2 through before 1.83.7 expose two MCP preview endpoints (POST /mcp-rest/test/connection and POST /mcp-rest/test/tools/list) th…KEVEPSS 93%analysed9.8CVE-2024-5751Litellm code injection vulnerabilityBerriAI/litellm version v1.35.8 contains a vulnerability where an attacker can achieve remote code execution. The vulnerability exists in the `add_de…EPSS 0.88%9.8CVE-2024-2952Litellm vulnerabilityBerriAI/litellm is vulnerable to Server-Side Template Injection (SSTI) via the `/completions` endpoint. The vulnerability arises from the `hf_chat_te…EPSS 1.3%9.5CVE-2026-49468Litellm authentication bypass by spoofing vulnerabilityLiteLLM is a proxy server (AI Gateway) to call LLM APIs in OpenAI (or native) format. Prior to 1.84.0, a Host-header parsing flaw in the LiteLLM prox…EPSS 3.0%9.4CVE-2026-35030Litellm improper authentication vulnerabilityLiteLLM 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…EPSS 0.88%

Source: NIST National Vulnerability Database (record CVE-2026-12797), CISA KEV, FIRST EPSS (scores of 2026-10-09). This page is refreshed as NVD updates the record.