Skip to main content
Press slash or control plus K to focus the search. Use the arrow keys to navigate results and press enter to open a threat.
Reconnecting to live updates…

CVE-2024-12911: CWE-89 Improper Neutralization of Special Elements used in an SQL Command in run-llama run-llama/llama_index

0
High
VulnerabilityCVE-2024-12911cvecve-2024-12911cwe-89
Published: Thu Mar 20 2025 (03/20/2025, 10:09:44 UTC)
Source: CVE Database V5
Vendor/Project: run-llama
Product: run-llama/llama_index

Description

A vulnerability in the `default_jsonalyzer` function of the `JSONalyzeQueryEngine` in the run-llama/llama_index repository allows for SQL injection via prompt injection. This can lead to arbitrary file creation and Denial-of-Service (DoS) attacks. The vulnerability affects the latest version and is fixed in version 0.5.1.

AI-Powered Analysis

AILast updated: 10/15/2025, 13:24:40 UTC

Technical Analysis

CVE-2024-12911 is a vulnerability classified under CWE-89 (Improper Neutralization of Special Elements used in an SQL Command), specifically an SQL injection flaw found in the default_jsonalyzer function of the JSONalyzeQueryEngine within the run-llama/llama_index repository. This vulnerability arises because the function fails to properly sanitize or neutralize special characters in user-supplied prompt inputs before incorporating them into SQL commands. As a result, an attacker can craft malicious prompts that inject arbitrary SQL commands, leading to unauthorized actions such as arbitrary file creation on the host system and Denial-of-Service (DoS) attacks by disrupting normal database operations. The vulnerability is remotely exploitable over the network without requiring authentication (AV:N/PR:N) but does require user interaction (UI:R) to trigger the injection. The scope is unchanged (S:U), meaning the impact is limited to the vulnerable component without affecting other system components. The CVSS v3.0 base score is 7.1, indicating a high severity level primarily due to the potential for availability disruption and integrity compromise. The vulnerability affects all versions prior to 0.5.1, where the issue has been addressed. No public exploits have been reported yet, but the nature of the flaw makes it a candidate for future exploitation, especially in environments where run-llama/llama_index is integrated into AI-driven data processing or query systems. The vulnerability highlights the risks of prompt injection attacks in AI frameworks that interface with databases without adequate input validation and sanitization.

Potential Impact

For European organizations, the impact of CVE-2024-12911 can be significant, especially for those relying on run-llama/llama_index for AI-driven data querying or processing. Successful exploitation can lead to arbitrary file creation, which may allow attackers to manipulate or plant malicious files within the system, potentially escalating to further compromise. The Denial-of-Service impact can disrupt critical business operations, causing downtime and loss of productivity. Since the vulnerability does not affect confidentiality directly, the primary concerns are integrity and availability. Organizations in sectors such as finance, healthcare, and critical infrastructure that utilize AI tools for data analysis are at higher risk. Additionally, the ease of remote exploitation without authentication increases the threat landscape, making exposed systems attractive targets. The lack of known exploits in the wild currently reduces immediate risk but does not eliminate the potential for future attacks. European entities with strict regulatory requirements around data integrity and service availability may face compliance and reputational risks if affected.

Mitigation Recommendations

To mitigate CVE-2024-12911, European organizations should immediately upgrade run-llama/llama_index to version 0.5.1 or later, where the vulnerability has been fixed. Beyond patching, organizations should implement strict input validation and sanitization on all user-supplied data, especially prompts that interact with SQL queries. Employ parameterized queries or prepared statements within the JSONalyzeQueryEngine to prevent injection attacks. Restrict network access to the vulnerable components by using firewalls and network segmentation to limit exposure. Monitor logs for unusual query patterns or errors indicative of injection attempts. Conduct regular security assessments and code reviews focusing on AI-related components that interface with databases. Additionally, implement runtime application self-protection (RASP) or web application firewalls (WAF) capable of detecting and blocking SQL injection attempts. Educate developers and users about the risks of prompt injection in AI systems to reduce inadvertent exposure.

Need more detailed analysis?Get Pro

Technical Details

Data Version
5.1
Assigner Short Name
@huntr_ai
Date Reserved
2024-12-24T07:51:29.340Z
Cvss Version
3.0
State
PUBLISHED

Threat ID: 68ef9b25178f764e1f470b1a

Added to database: 10/15/2025, 1:01:25 PM

Last enriched: 10/15/2025, 1:24:40 PM

Last updated: 10/16/2025, 12:42:05 PM

Views: 1

Community Reviews

0 reviews

Crowdsource mitigation strategies, share intel context, and vote on the most helpful responses. Sign in to add your voice and help keep defenders ahead.

Sort by
Loading community insights…

Want to contribute mitigation steps or threat intel context? Sign in or create an account to join the community discussion.

Actions

PRO

Updates to AI analysis require Pro Console access. Upgrade inside Console → Billing.

Please log in to the Console to use AI analysis features.

Need enhanced features?

Contact root@offseq.com for Pro access with improved analysis and higher rate limits.

Latest Threats