CVE-2024-6587Public exploitNetwork attack vector

API-key theft via caller-controlled provider URL

Published Sep 13, 2024 · Updated Sep 13, 2024

SSRF in Berri AI LiteLLM before 1.44.9 allows remote attackers to steal an OpenAI API key through a crafted chat request. The proxy's POST /chat/completions handler accepts api_base or base_url from the request body and uses it as the provider destination without restricting caller-controlled routing. Unauthenticated network access to that endpoint is sufficient, and the outbound request discloses the configured OpenAI API key to the chosen host.

CVSS severity8.7
High
EPSS probability35.32%
Next 30 days · Sep 16, 2026
Known exploitationUnconfirmed
Based on sourced intelligence
Hinoki checkNot available
Coverage for this vulnerability

See if you're affected

Explore vulnerability checks for your environment with Hinoki.

Book a demo

Summary

What happened

SSRF in Berri AI LiteLLM before 1.44.9 allows remote attackers to steal an OpenAI API key through a crafted chat request. The proxy's POST /chat/completions handler accepts api_base or base_url from the request body and uses it as the provider destination without restricting caller-controlled routing. Unauthenticated network access to that endpoint is sufficient, and the outbound request discloses the configured OpenAI API key to the chosen host.

The record

CVE
CVE-2024-6587
Published
Sep 13, 2024
Updated
Sep 13, 2024
Vendor
Berri AI
Product
LiteLLM
Classifications
CWE-918, T1190
Attack vector
network
Privileges
unauthenticated

Timeline

How it unfolded

  1. Sep 13, 2024CVE publishedPublication date reported by the CVE source.
  2. Sep 13, 2024Record updatedLatest update available in the CVE record.

Exploitability

Present is not the same as exploitable

Compare your product and version with the public record. A matching version still requires validation against your environment.

Is a vulnerable build present?

Compare these published version ranges with your installed build and any vendor patches.

  1. Affected versionversion=0 <1.44.9
  2. Affected versionversion=unspecified <1.44.9

What conditions does exploitation require?

Attack vectornetwork
Required privilegesunauthenticated

What is affected?

Berri AI · LiteLLMversion=0 <1.44.9; version=unspecified <1.44.9

Published CVSS scores

7.5GitLab Advisory DatabaseCVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:N/A:N
8.7GitHub Advisory DatabaseCVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:H/VI:N/VA:N/SC:N/SI:N/SA:N

CVSS describes severity. EPSS estimates exploitation probability.

Attacks

What attackers are doing with it

Daily unique IPs observed by Shadowserver honeypots for known exploited vulnerabilities (KEVs). Missing observations do not establish an absence of attacks.

Daily unique IPsNo honeypot observations are available for this CVE in the selected window.

No observations available

Sep 10, 2026Sep 16, 2026
Latest reporting daySep 16, 2026
Latest daily unique IPsUnavailable
Prior 30-day averageUnavailable
SourceShadowserver honeypots (KEV)
Vectornetwork
Privilegesunauthenticated
Known exploitationUnconfirmed
Public exploitPublished

Weakness, pattern, technique

CWE-918Server-Side Request Forgery (SSRF)
T1190Exploit Public-Facing Application

Public exploit references

Labels summarize the accepted research assessment. They do not indicate a test against your environment.

Technologies

Your stack

See the directory against your own environment.

Your stack

Check the software in your environment

Book a demo to see how Hinoki identifies affected software and validates exploitability in your environment.

Book a demo