SQL Injection in Spring AI CosmosDB Vector Store

SQL injection vulnerability in Spring AI's `CosmosDBVectorStore` allows attackers to execute arbitrary SQL queries via crafted document IDs. Affected versions: Spring AI: 1.0.0 - 1.0.5 (fixed in 1.0.6), 1.1.0 - 1.1.4 (fixed in 1.1.5)

Casky was already ahead

This CVE exploits attack patterns that Casky's 301matched skills already investigate — long before this vulnerability was disclosed. Claude's reasoning model maps these techniques to MITRE ATT&CK, so practitioners who ran these skills have already seen the threat behaviour in their findings.

Analysis

Spring AI versions 1.0.0-1.0.5 and 1.1.0-1.1.4 contain a critical SQL injection vulnerability in the CosmosDBVectorStore component that allows attackers to execute arbitrary SQL queries by manipulating document IDs. This vulnerability (CWE-89) is particularly dangerous because it affects applications using Spring AI for vector database operations, which are increasingly common in AI-powered applications handling sensitive data. Organizations running affected versions in production environments face immediate risk of unauthorized data access, modification, or exfiltration. The vulnerability requires no authentication and can be exploited remotely, making it a high-priority patch for any team leveraging Spring AI's vector storage capabilities.

Casky's security skills detect the attack patterns associated with this vulnerability by identifying the MITRE ATT&CK techniques at play: T1190 (Exploit Public-Facing Application) and T1059 (Command and Scripting Interpreter). Using Claude AI's extended reasoning capabilities, Casky practitioners would observe SQL syntax anomalies in document ID parameters—such as quotes, semicolons, UNION statements, or comment sequences—that indicate injection attempts. The platform's 754 mapped security skills enable detection of malicious query construction patterns before they reach the database layer, alerting teams to suspicious document ID formatting that deviates from expected input. Practitioners reviewing findings would see flagged instances where attacker-controlled input flows directly into SQL query construction, coupled with evidence of attempted command execution through database-level operations.

Live Threat IntelligenceComing soon

Casky Risk Score

—

EPSS Probability

—

Shodan Exposure

—

GreyNoise Signal

—

Composite risk scoring from EPSS, CISA KEV, Shodan, and GreyNoise — 21 security APIs correlated into a single Casky Risk Score. Coming in Casky Pro. Join early access →

301 Casky skills cover this attack pattern

These skills use Claude AI's reasoning model to surface findings in the same attack categories as CVE-2026-40978.

analyzing-apt-group-with-mitre-navigator

threat intelligence · low

Run

analyzing-campaign-attribution-evidence

threat intelligence · low

Run

analyzing-certificate-transparency-for-phishing

threat intelligence · low

Run

MITRE ATT&CK Techniques

TA0001 TA0043

Investigate the attack patterns behind CVE-2026-40978

Casky has 301 skills that investigate the attack patterns behind CVE-2026-40978. Run one and get CVSS-scored findings in 3 minutes.

Run the skill that detects this →

X Instagram LinkedIn