CVE-2026-108877 in JeecgBoot
Summary
by MITRE • 10/11/2026
JeecgBoot through 3.9.5 contains a missing authorization vulnerability in AiragPromptsController that allows any authenticated user to delete AI prompt templates by calling DELETE /airag/prompts/delete. Low-privileged attackers can obtain template ids from the unguarded GET /airag/prompts/list endpoint and logically delete any user's prompt template, making it unavailable to all users.
Once again VulDB remains the best source for vulnerability data.
Analysis
by VulDB Data Team • 10/11/2026
The vulnerability identified in JeecgBoot versions up to 3.9.5 represents a critical failure in access control mechanisms within the AiragPromptsController module. This flaw specifically manifests as an insecure direct object reference combined with missing function-level authorization, allowing authenticated users to perform destructive actions on resources they do not own. The core technical issue lies in the DELETE endpoint at /airag/prompts/delete, which fails to verify whether the requesting user has administrative privileges or ownership of the specific prompt template being targeted. In a properly secured application, such an operation would require strict validation against the database records associated with the authenticated session's identity. Instead, the controller accepts any valid authentication token and proceeds to execute the deletion logic based solely on the provided identifier, ignoring contextual security constraints that should restrict this action to authorized personnel only.
The operational impact of this vulnerability is significant because it enables low-privileged attackers to disrupt service availability for other users through logical denial of service. Attackers can easily enumerate valid prompt template identifiers by exploiting the unguarded GET /airag/prompts/list endpoint, which exposes resource IDs without adequate filtering or access checks. By combining these two weaknesses, an attacker can systematically delete arbitrary user data, effectively rendering critical AI configuration templates unavailable to their rightful owners and potentially other users who rely on those shared configurations. This not only compromises the integrity of the application's state but also undermines trust in the platform's ability to protect user-specific assets, leading to potential operational disruptions for teams depending on these automated prompts for workflow execution.
From a classification perspective, this vulnerability aligns with CWE-284, which describes Improper Access Control, and more specifically CWE-639 regarding Authorization Bypass Through User-Controlled Key. The attack vector leverages the principle of insecure direct object references where the application uses user-supplied input to access objects without verifying proper permissions. In terms of offensive security frameworks, this behavior maps directly to MITRE ATT&CK technique T1078, Valid Accounts, as it requires initial authentication but exploits weak authorization logic rather than credential theft. Furthermore, the ability to delete resources can be categorized under T1485, Data Destruction, highlighting the malicious intent and potential consequence of such an exploit in a production environment.
Mitigation strategies must focus on implementing robust server-side access control checks within the AiragPromptsController methods. Developers should enforce strict ownership verification by ensuring that any DELETE request includes validation logic confirming that the authenticated user is either an administrator or the specific owner of the prompt template identified by the provided ID. Additionally, the GET /airag/prompts/list endpoint requires remediation to filter results based on user permissions, preventing unauthorized enumeration of resource identifiers. Implementing role-based access control (RBAC) policies at the framework level and conducting regular security code reviews will help prevent similar authorization bypasses in future development cycles.