Kong Gateway Enterprise and AI Gateway: SQL injection in LLM usage counters
Summary
Kong Gateway Enterprise and Kong AI Gateway contain an SQL injection flaw (CWE-89) in the LLM usage counters. A client can send a crafted model name and run arbitrary SQL against the Kong database. The CVSS 4.0 base score is 8.7 (High), and exploitation requires network access and a valid low-privilege client or consumer context.
Affected
- Kong Gateway Enterprise 3.10.0.0 to before 3.10.0.20
- Kong Gateway Enterprise 3.13.0.0 to before 3.13.0.12
- Kong Gateway Enterprise 3.14.0.0 to before 3.14.0.17
- Kong Gateway Enterprise 3.15.0.0 to before 3.15.0.7
- Kong Gateway Enterprise 3.16.0.0 to before 3.16.0.2
- Kong AI Gateway 2.2.0 to before 2.2.1
Fixed in
- Kong Gateway Enterprise 3.10.0.20, 3.13.0.12, 3.14.0.17, 3.15.0.7 and 3.16.0.2 (per the CNA record; the 3.16.0.2 changelog entry confirms the fix)
- Kong AI Gateway 2.2.1 (per the CNA record; unverified, see note below)
- Note: The AI Gateway changelog shows no fix entry for this CVE in the excerpt reviewed. Research notes point to a 2.3.0 heading, which could not be confirmed. Check the AI Gateway changelog before planning the upgrade.
Workaround
Unknown. Kong has not published a workaround. A third-party tracker suggests avoiding untrusted input for model names and adding input validation; this is not vendor guidance and has not been verified.
Exploitation
No evidence of exploitation in the wild was found. The CVE is not listed as known exploited (KEV: No) in the sources reviewed. No public proof of concept was found, which does not prove that none exists.
Our recommendation
- Inventory all Kong Gateway Enterprise and Kong AI Gateway instances and their versions.
- Upgrade Kong Gateway Enterprise to the fixed version of your branch. Confirm the correct fixed version for Kong AI Gateway before upgrading.
- Until patched, restrict which clients can send requests to the gateway and review logs for unusual model names.
- Re-check NVD and the Kong changelogs once NVD analysis is complete, and check for any KEV listing.
Public exploits
PoC-in-GitHub, Metasploit and Exploit-DB list nothing for this CVE.
Research sources
- CVE Record CVE-2026-92544 www.cve.org
- Kong Gateway changelog developer.konghq.com
- Kong AI Gateway changelog developer.konghq.com
- Strix AI CVE-2026-92544 www.strix.ai
- Threat Intelligence CVE-2026-92544 cve.threatint.com
- OffSeq Radar CVE-2026-92544 radar.offseq.com
Researched and written up in English and German by an AI (Claude by Anthropic) based on NVD, CISA, EPSS and the sources listed. The vendor information is authoritative. Researched on 11 Nov 2026, 00:57.
Original description (NVD)
SQL injection in the LLM usage counters of Kong Gateway Enterprise and Kong AI Gateway may allow a client to execute arbitrary SQL against the Kong database via a crafted model name.
Sources
The tracker combines official, freely usable data sources – updated continuously:
- NVD (NIST) – CVEs, rating (CVSS), affected products
- CISA KEV – vulnerabilities confirmed by CISA as actively exploited
- CISA Vulnrichment – CISA assessment (SSVC) in the official CVE record
- FIRST EPSS – probability of exploitation
- Microsoft MSRC – Patch Tuesday: vulnerabilities, severity, updates
- PoC-in-GitHub – public proof-of-concept repositories
- Metasploit (Rapid7) – modules of the Metasploit Framework
- Exploit-DB (OffSec) – archive of public exploits (links only)
- Patch days: official vendor pages (linked per vendor) (Vendor patch days)
This product uses data from the NVD API but is not endorsed or certified by the NVD. EPSS data © FIRST.org.
Affected? We can help.
We check whether your systems are affected, install updates and close the gap – continuously with the System Health Checker if you wish.