CISA Adds One Known Exploited Vulnerability to Catalog

What Happened

CISA has identified that a vulnerability in a software called BerriAI LiteLLM is actively being used by attackers. This flaw allows attackers to manipulate databases through a technique called SQL Injection. Because it is actively being exploited, it poses a significant risk to organizations. All organizations, especially federal agencies, should update or patch this software immediately to protect their networks.

Key Takeaways

• CISA has added CVE-2026-42208 to the Known Exploited Vulnerabilities (KEV) Catalog.
• The vulnerability is identified as a SQL Injection flaw in BerriAI LiteLLM.
• There is confirmed evidence of active exploitation of this vulnerability in the wild.
• Federal Civilian Executive Branch (FCEB) agencies are mandated to remediate this vulnerability under BOD 22-01.

Affected Systems

• BerriAI LiteLLM

Vulnerabilities (CVEs)

• CVE-2026-42208

Attack Chain

Threat actors are actively exploiting CVE-2026-42208, a SQL injection vulnerability in BerriAI LiteLLM. Successful exploitation likely allows attackers to manipulate backend database queries via crafted input, potentially leading to unauthorized data access, data modification, or further compromise of the underlying system.

Detection Availability

• YARA Rules: No
• Sigma Rules: No
• Snort/Suricata Rules: No
• KQL Queries: No
• Splunk SPL Queries: No
• EQL Queries: No
• Other Detection Logic: No

No specific detection rules or queries are provided in the source article.

Detection Engineering Assessment

EDR Visibility: Low — SQL injection exploitation occurs at the application and database layer, which is typically not directly visible to standard EDR without specific application log integration. Network Visibility: Medium — WAFs and network intrusion detection systems may catch anomalous SQL syntax in web requests targeting the LiteLLM application. Detection Difficulty: Moderate — Detecting SQL injection requires proper logging of web requests and database queries, as well as the ability to distinguish malicious payloads from legitimate application traffic.

Required Log Sources

• Web Application Firewall (WAF) logs
• Web server access logs
• Database query logs

Hunting Hypotheses

Hypothesis	Telemetry	ATT&CK Stage	FP Risk
Look for anomalous SQL syntax or unexpected database queries originating from the BerriAI LiteLLM application process.	Database query logs, Web server access logs	Initial Access	Medium

Control Gaps

• Lack of WAF rules for specific LiteLLM endpoints
• Insufficient database query monitoring

Key Behavioral Indicators

• Unexpected SQL keywords in HTTP request parameters
• Database errors returned to the web client

False Positive Assessment

• Low

Recommendations

Immediate Mitigation

• Apply the latest vendor patches or updates for BerriAI LiteLLM to remediate CVE-2026-42208.

Infrastructure Hardening

• Deploy and configure Web Application Firewalls (WAF) to inspect and block malicious SQL injection payloads.
• Implement the principle of least privilege for the database user account used by the LiteLLM application.

User Protection

• N/A

Security Awareness

• Ensure vulnerability management teams prioritize alerts and required remediations from the CISA KEV catalog.

MITRE ATT&CK Mapping

• T1190 - Exploit Public-Facing Application
• T1059 - Command and Scripting Interpreter