CVE-2025-27135
Published: 25 February 2025
Description
Adversaries may destroy data and files on specific systems or in large numbers on a network to interrupt availability to systems, services, and network resources.
Security Summary
CVE-2025-27135 is a SQL injection vulnerability (CWE-89) in RAGFlow, an open-source Retrieval-Augmented Generation (RAG) engine. Versions 0.15.1 and prior are affected, specifically the ExeSQL component, which extracts SQL statements directly from input and executes them on the database without sanitization. The vulnerability was published on 2025-02-25 and carries a CVSS v3.1 base score of 9.8 (AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H), indicating critical severity.
An unauthenticated remote attacker with network access can exploit this vulnerability through low-complexity attacks requiring no user interaction. Exploitation allows arbitrary SQL query execution, enabling high-impact compromise of confidentiality, integrity, and availability, such as data exfiltration, modification, or deletion from the underlying database.
As of publication, no patched version of RAGFlow is available. Relevant advisories and details are documented in the GitHub security advisory (GHSA-3gqj-66qm-25jq), the affected ExeSQL source code, and external analyses on provided Notion pages.
RAGFlow's role as a RAG engine introduces AI/ML relevance, as deployments in LLM pipelines could expose sensitive data stores to remote compromise. No real-world exploitation has been reported in the available information.
Details
- CWE(s)
Affected Products
MITRE ATT&CK Enterprise Techniques
Why these techniques?
SQL injection in public-facing RAGFlow ExeSQL allows remote unauthenticated arbitrary SQL execution, directly enabling T1190 (Exploit Public-Facing Application) for initial access and facilitating T1213.006 (Databases) for data collection, T1565.001 (Stored Data Manipulation) for integrity impacts, and T1485 (Data Destruction) for availability impacts via the database.