Skip to content
Server Security Center
KEVcritical9.3

CVE-2026-42208: BerriAI LiteLLM SQL Injection

BerriAI LiteLLM · listed in the CISA catalog of actively exploited vulnerabilities since 2026-05-08.

CVSS v4.0 (CVE record)
9.3
critical
EPSS (exploit probability, 30 days)
89%
higher than 99.7% of all CVEs
Added to CISA KEV
2026-05-08
US federal due date 2026-05-11
Automatable (CISA SSVC)
yes
technical impact: total

CVE-2026-42208 explained

CVE-2026-42208 is a SQL injection vulnerability in BerriAI LiteLLM, a proxy server (AI Gateway) for calling LLM APIs. During proxy API key checks, a caller-supplied key value was mixed into database query text. Unauthenticated attackers can reach it, and CISA lists it as known to be exploited.

Impact

An unauthenticated attacker can send a crafted Authorization header to any LLM API route. This lets them read data from the proxy's database and possibly modify it. The result can be unauthorised access to the proxy and the credentials it manages. CISA rates it automatable, so it can be exploited at scale, with total technical impact. Attackers are known to use this vulnerability.

Who is affected

BerriAI litellm is affected from 1.81.16 up to but not including 1.83.7. The description states that the issue was patched in version 1.83.7. However, the structured fixedIn field of the CVE record names no fixed version. Confirm the fixed release in the vendor advisory before upgrading.

Written with AI from the data on this page and checked against it automatically.

Official description

LiteLLM is a proxy server (AI Gateway) to call LLM APIs in OpenAI (or native) format. From version 1.81.16 to before version 1.83.7, a database query used during proxy API key checks mixed the caller-supplied key value into the query text instead of passing it as a separate parameter.

An unauthenticated attacker could send a specially crafted Authorization header to any LLM API route (for example POST /chat/completions) and reach this query through the proxy's error-handling path. An attacker could read data from the proxy's database and may be able to modify it, leading to unauthorised access to the proxy and the credentials it manages.

This issue has been patched in version 1.83.7.

CISA required action: Apply mitigations per vendor instructions, follow applicable BOD 22-01 guidance for cloud services, or discontinue use of the product if mitigations are unavailable.

CISA notes

Affected and fixed versions

From the CVE record published by the vendor or CVE numbering authority.

  • BerriAI litellm
    Affected
    • ≥ 1.81.16
    • < 1.83.7
    Fixed in
    not stated

How to fix it

  • The CVE record names no fixed version – check the vendor advisory and CISA's required action for updates and mitigations.
  • Vendor advisory: github.com/BerriAI/litellm/security/advisories/GHSA-r75f-…
  • CISA's deadline for US federal agencies: 2026-05-11 (passed) – a useful benchmark for everyone else, too.

What to do

  1. Upgrade LiteLLM to 1.83.7 or later, as stated in the description, after confirming it in the vendor advisory.
  2. Apply the vendor's mitigations and follow CISA's required action, including BOD 22-01 guidance for cloud services.
  3. Restrict network access to the LiteLLM proxy so its API routes are not exposed to untrusted networks.
  4. Check proxy logs and the database for signs of compromise, such as unusual Authorization headers or unexpected data changes.
  5. Rotate API keys and provider credentials managed by the proxy if compromise is suspected.

The CVE record's structured data names no fixed version, but the description states the issue was patched in 1.83.7. Confirm this in the vendor advisory. Until you can upgrade, apply the vendor's mitigations and follow CISA's required action. If mitigations are unavailable, CISA says to discontinue use of the product. Limit network exposure of the proxy in the meantime.

Written with AI from the data on this page and checked against it automatically.

CISA's assessment (SSVC)

Exploitation
Exploited in the wild
Automatable
Yes – attackers can automate it at scale
Technical impact
Total – full control of the component or total information disclosure

Weakness: CWE-89 Improper Neutralization of Special Elements used in an SQL Command ('SQL Injection')

Questions about CVE-2026-42208

Does an attacker need valid credentials?
No. The description states that an unauthenticated attacker can trigger the flaw by sending a crafted Authorization header to any LLM API route.
Is this vulnerability being exploited?
Yes. CISA added CVE-2026-42208 to its Known Exploited Vulnerabilities catalog on 2026-05-08, with a US federal due date of 2026-05-11. Ransomware use is listed as unknown.
What data is at risk?
The attacker can read the proxy's database and may be able to modify it. This can expose the proxy and the credentials it manages, so consider rotating managed keys after upgrading.

Written with AI from the data on this page and checked against it automatically.

References

Data: CISA Known Exploited Vulnerabilities (public domain) · CVE Program records incl. CISA ADP enrichment · OSV.dev (CC BY 4.0) incl. Ubuntu and Debian security data · FIRST EPSS · endoflife.date. Updated automatically – facts last changed 2026-11-21. Always confirm with the vendor's or distribution's advisory before patching.