CVE-2026-76543 in Splunk
Summary
by MITRE • 11/20/2026
In Splunk Enterprise versions below 10.6.1, 10.4.5, 10.2.9, 10.0.12, and 9.4.17, a user who holds the "power" Splunk role could store attacker-controlled Search Processing Language (SPL) in a shared data model. The vulnerability is possible because the Data Model Editor does not properly neutralize data model field names when it builds a preview search. The vulnerability requires the attacker to phish the affected user by tricking them into initiating a request within their browser. The attacker should not be able to exploit the vulnerability at will. For more information see Add a regular expression field (https://help.splunk.com/en/splunk-enterprise/manage-knowledge-objects/knowledge-management-manual/10.4/define-data-model-dataset-fields/add-a-regular-expression-field), Define roles on the Splunk platform with capabilities (https://help.splunk.com/en/splunk-enterprise/administer/manage-users-and-security/10.4/manage-splunk-platform-users-and-roles/define-roles-on-the-splunk-platform-with-capabilities), and About the search language (https://help.splunk.com/en/splunk-enterprise/search/search-manual/10.4/search-overview/about-the-search-language) in the Splunk documentation.
If you want to get best quality of vulnerability data, you may have to visit VulDB.
Analysis
by VulDB Data Team • 11/20/2026
In versions of Splunk Enterprise prior to 10.6.1, 10.4.5, 10.2.9, 10.0.12, and 9.4.17, a critical security flaw exists within the Data Model Editor that allows for the storage of attacker-controlled Search Processing Language commands in shared data models. This vulnerability stems from an insufficient neutralization of user-supplied input when constructing preview searches, specifically regarding field names defined via regular expressions. The core technical failure lies in the application's inability to properly sanitize or validate these field name inputs before they are processed as part of a search query structure. Because the Data Model Editor integrates directly with the underlying search engine to generate previews for validation purposes, any malicious code embedded within the field definition is executed during this preview phase rather than being treated purely as metadata. This architectural design choice creates an execution path where untrusted data from a user interface form can be promoted into executable commands without adequate security checks or context switching that would typically prevent such actions in standard search operations.
The operational impact of this vulnerability centers on the potential for unauthorized code execution within the Splunk environment, contingent upon specific privilege levels and social engineering tactics. An attacker must first compromise or trick a user who possesses at least the "power" role to initiate the malicious request through their browser. The power role is significant because it grants users substantial capabilities within the platform, including the ability to create and modify knowledge objects such as data models. Once this privileged user interacts with a crafted payload by entering attacker-controlled SPL into the regular expression field definition, the system processes the input during the preview generation phase. This interaction effectively allows an authenticated user to inject arbitrary commands that are then executed in the context of their own privileges or potentially escalated depending on how the search is subsequently utilized within shared dashboards or reports. The requirement for phishing indicates that this vulnerability cannot be exploited remotely by unauthenticated users, but it poses a severe risk to organizations where insider threats or compromised credentials with moderate privilege levels exist.
From a classification perspective, this flaw aligns closely with CWE-94 Improper Control of Generation of Code (Code Injection), as the application fails to properly neutralize user input before incorporating it into code that is executed by the system. Furthermore, the exploitation technique maps to MITRE ATT&CK techniques involving Command and Scripting Interpreter abuse, specifically leveraging legitimate administrative tools for malicious purposes. The attack vector relies on social engineering to bypass authentication controls, which relates to initial access vectors where trusted users are manipulated into performing actions that compromise integrity. The fact that the vulnerability requires user interaction via a browser highlights the importance of secure coding practices in web-based management interfaces and underscores the risks associated with rich client-side interactions that trigger server-side processing without rigorous input validation.
Mitigation strategies for this vulnerability involve immediate upgrading to Splunk Enterprise version 10.6.1 or later, where these specific versions have addressed the sanitization issues within the Data Model Editor. For organizations unable to upgrade immediately due to operational constraints, restricting access to data model creation and modification capabilities is essential. Administrators should audit role definitions to ensure that only highly trusted individuals hold permissions to edit shared knowledge objects. Additionally implementing strict input validation rules at the application layer for any fields that are subsequently used in search construction can provide a defensive layer. Security teams should also monitor logs for unusual patterns of data model edits or preview searches originating from unexpected sources, as these may indicate attempted exploitation activities. Regular security training focused on phishing awareness is equally critical to prevent users with elevated privileges from being tricked into initiating malicious requests that could lead to broader system compromise.