Advisory record · GHSA-2jxw-4hm4-6w87

LlamaIndex (aka llama_index) through 0.9.34 allows SQL injection via the Text-to-SQL feature in NLSQLTableQueryEngine, SQLTableRetrieverQueryEngine, NLSQLRetriever, RetrieverQueryEngine, and PGVect...

CVSS 9.8 CriticalCriticalEPSS 0.0065 · 52nd percentile

Critical, because:

  • CVSS score 9.8: 9.0 or more.

How we decide what’s critical

Published by the GitHub Advisory Database at 22 Jan 2024, 01:15 UTC. agentprime has not analysed this yet, so whether it affects any app is not known.

PackageEcosystemAffectedFixed in
llama-indexPyPI<0.9.350.9.35
Identifiers
GHSA-2jxw-4hm4-6w87 · CVE-2024-23751 · PYSEC-2024-12
Weakness (CWE)
CWE-89
Published
22 Jan 2024
Updated
10 Jun 2026

Machine-written summary

Summary written by Claude Haiku from the source advisory. Not reviewed by a person.

LlamaIndex (llama-index) versions before 0.9.35 are vulnerable to SQL injection through the Text-to-SQL features in several query engines and retrievers, including NLSQLTableQueryEngine and PGVectorSQLQueryEngine. An attacker who can submit natural-language input to these features may be able to cause arbitrary SQL to run, such as deleting or dropping database tables. Upgrade llama-index to version 0.9.35 or later.

Written 10 Nov 2026.

Not analysed yet

We haven’t published our analysis of this Advisory yet. Once a person has reviewed it, this page will say what’s vulnerable, what has to be true for your app to be affected, and how to check.

Severity and exploitation

ScoreRatingCVSSVectorSource
9.8 Critical 3.1 CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H ADP (CISA-ADP)
  • CISA KEV Not listed.
  • EPSS 0.0065, the 52nd percentile, on 21 Nov 2026: FIRST’s estimate of the chance of exploitation in the next 30 days.

Timeline

This record came from the full history of the GitHub Advisory Database, published 22 Jan 2024. New records show when each source confirmed them.

Sources

Facts on this page come from the GitHub Advisory Database (CC-BY 4.0), OSV and the other sources listed, as published. We don’t reproduce full advisory descriptions, which can contain exploit code.