CVE-2026-33324: SQLBot prompt injection allows arbitrary SQL execution and remote code execution
SQLBot is an intelligent Text-to-SQL system based on large language models and RAG. In versions 1.7.0 and earlier, the Text2SQL chat interface is vulnerable to prompt injection. The user-provided question parameter is directly concatenated into the LLM prompt without filtering or escaping, and the SQL extracted from the LLM response is executed against the database without validation or sanitization. An authenticated attacker can craft a malicious question to manipulate the LLM into generating and executing arbitrary SQL statements. When connected to a PostgreSQL data source, this can lead to remote code execution via COPY FROM PROGRAM. This issue has been fixed in version 1.7.1.
Affected Software
Remediation
Recommended actions to resolve this vulnerability, in priority order.
- Upgrade
Upgrade
SQLBot (Text2SQL chat interface)to a version that resolves this vulnerability.Fixed in 1.7.1 - Compensating control
If SQLBot is connected to a PostgreSQL data source, restrict the database capability to prevent COPY FROM PROGRAM (e.g., revoke/limit permissions for PROGRAM/COPY FROM PROGRAM) to mitigate remote code execution risks.
Event History
Frequently Asked Questions
What is the severity of CVE-2026-33324?
CVE-2026-33324 is classified as a critical vulnerability due to its potential for arbitrary SQL execution and remote code execution.
How do I fix CVE-2026-33324?
To mitigate CVE-2026-33324, upgrade SQLBot to version 1.7.1 or later, which addresses the prompt injection vulnerability.
What systems are affected by CVE-2026-33324?
CVE-2026-33324 affects SQLBot versions 1.7.0 and earlier.
What type of vulnerability is CVE-2026-33324?
CVE-2026-33324 is a prompt injection vulnerability in the Text-to-SQL system of SQLBot.
Can CVE-2026-33324 lead to data breaches?
Yes, CVE-2026-33324 can lead to data breaches as it allows for arbitrary SQL queries that may compromise sensitive data.