Executive brief
JeecgBoot is a low-code development platform used to rapidly build enterprise applications. A security vulnerability in its AI model configuration component allows an authenticated user to force the server to make unauthorized requests to internal or external network addresses. This could allow an attacker to scan internal corporate networks, access sensitive cloud metadata, or bypass firewalls to reach systems not otherwise exposed to the internet.
Technical details
A Server-Side Request Forgery (SSRF) vulnerability exists in the /airag/airagModel/test endpoint of JeecgBoot. The root cause is the lack of validation or sanitization of the 'baseUrl' argument within the AiragModel request body before it is passed to the langchain4j framework to initiate outbound HTTP requests. An authenticated attacker can provide a malicious URL to probe internal network services, perform port scanning, or access sensitive cloud instance metadata (e.g., AWS/GCP metadata endpoints). Additionally, the endpoint lacks permission annotations, allowing any authenticated user to trigger the flaw, and the malicious configuration is persisted to the database, enabling stored SSRF. The vulnerability affects AiragModelController.java, AIChatHandler.java, and EmbeddingHandler.java.
Affected products
- JeecgBoot JeecgBoot up to 3.9.2
Timeline
- 2026-04-30: other: Version 3.9.2 released
- 2026-05-06: disclosed: Issue reported on GitHub repository
- 2026-06-01: advisory: CVE published and NVD record created