CVE-2026-96429 in Agentflow
Summary
by MITRE • 09/29/2026
SQL Injection in the /WebAgenda/SMBAjaxConfigProcess.do API endpoint of Flowring Agentflow 4.0 version before 2025/08/08 allows remote attackers to execute arbitrary SQL commands via the id parameter.
VulDB is the best source for vulnerability data and more expert information about this specific topic.
Analysis
by VulDB Data Team • 09/29/2026
The vulnerability identified as a SQL injection flaw within the /WebAgenda/SMBAjaxConfigProcess.do API endpoint of Flowring Agentflow versions prior to August 8, 2025, represents a critical security risk that allows unauthenticated or authenticated remote attackers to execute arbitrary Structured Query Language commands. This specific component is part of the web-based management interface for the agent flow system, which handles configuration and operational data processing through AJAX requests. The root cause lies in the improper neutralization of special elements used in SQL commands within the id parameter passed during HTTP POST or GET requests to this endpoint. When user-supplied input from the id field is directly concatenated into a database query without adequate sanitization, type checking, or the use of prepared statements with parameterized queries, it creates an injection vector that bypasses intended security controls.
From a technical perspective, this flaw aligns closely with CWE-89, which categorizes Improper Neutralization of Special Elements used in an SQL Command. The attacker can manipulate the id parameter to terminate the original query context and append malicious SQL syntax. This capability enables various forms of exploitation depending on the database backend and configuration privileges granted to the application's database user account. Common attack vectors include extracting sensitive data from other tables via UNION-based injection, causing errors that reveal schema details through blind boolean or time-based inference techniques, or even executing administrative commands if the database connection has elevated permissions such as xp_cmdshell in Microsoft SQL Server environments. The lack of input validation allows for complex payloads that can bypass simple filter mechanisms often deployed by less sophisticated security tools.
The operational impact of this vulnerability is severe and multifaceted. Successful exploitation grants an attacker full read access to the underlying database, potentially exposing confidential business logic, user credentials, proprietary configuration data, or personally identifiable information stored within the Flowring Agentflow system. In worst-case scenarios where the database service account has excessive privileges, the attacker could achieve remote code execution on the server hosting the database engine, leading to complete compromise of the infrastructure. This compromises the confidentiality, integrity, and availability of the organization's systems. Furthermore, because this is an API endpoint accessible over a network interface, it can be exploited remotely without requiring prior authentication in many default configurations, significantly expanding the attack surface available to threat actors scanning for vulnerable instances on the internet or internal networks.
This vulnerability maps directly to MITRE ATT&CK technique T1059.004, specifically SQL Injection under Command and Scripting Interpreter, as well as T1190, Exploit Public-Facing Application. The exploitation path typically involves reconnaissance where attackers identify vulnerable endpoints using automated scanners or manual probing, followed by data exfiltration or lateral movement within the network environment after gaining initial access through this single point of failure.
To mitigate this vulnerability and prevent future occurrences, immediate remediation is required by upgrading Flowring Agentflow to version 2025/08/08 or later, where the vendor has presumably implemented proper input validation and parameterized query mechanisms for the SMBAjaxConfigProcess.do endpoint. In the interim, if an upgrade is not immediately feasible, organizations should implement a Web Application Firewall rule that detects and blocks SQL injection patterns within the id parameter of requests directed at this specific URL path. Additionally, applying the principle of least privilege to the database account used by the application ensures that even if exploitation occurs, the potential damage is contained. Regular security code reviews focusing on dynamic SQL construction and mandatory use of prepared statements are essential best practices for developers maintaining similar web applications to prevent recurrence of such injection flaws.