Arbitrary SQL execution via unchecked LLM-generated statements
Published Jan 22, 2024 · Updated Jun 20, 2025
SQL injection in LlamaIndex 0.9.34 and earlier allows attackers to execute arbitrary SQL through natural-language Text-to-SQL input. The five affected query-engine and retriever classes send LLM-generated SQL to the configured database without checking whether the statement is read-only. Exploitation requires an application to pass attacker-controlled prompts to an affected Text-to-SQL component, and database permissions bound the ability to read, change, or delete data.
Summary
What happened
SQL injection in LlamaIndex 0.9.34 and earlier allows attackers to execute arbitrary SQL through natural-language Text-to-SQL input. The five affected query-engine and retriever classes send LLM-generated SQL to the configured database without checking whether the statement is read-only. Exploitation requires an application to pass attacker-controlled prompts to an affected Text-to-SQL component, and database permissions bound the ability to read, change, or delete data.
The record
- CVE
- CVE-2024-23751
- Published
- Jan 22, 2024
- Updated
- Jun 20, 2025
- Vendor
- Unknown vendor
- Product
- Unknown product
- Classifications
- CWE-89, T1190
- Attack vector
- network
- Privileges
- unauthenticated
Timeline
How it unfolded
- Jan 22, 2024CVE publishedPublication date reported by the CVE source.
- Jun 20, 2025Record 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?
What conditions does exploitation require?
What is affected?
Affected products and versions are unavailable in this record.
Published CVSS scores
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.
Weakness, pattern, technique
Public exploit references
- Text-to-SQL DROP TABLE proof of conceptproof of concept · demonstrated
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