CVE-2026-31020: DocsGPT vulnerability
In DocsGPT 0.15.0 and below, the application provides a custom prompt feature that allows users to define prompt content used during chatbot interactions. This functionality renders user-supplied prompt data using Jinja templates without input sanitization or sandboxing. An unauthenticated attacker can inject malicious template expressions, leading to a server-side template injection (SSTI) vulnerability that can be exploited to achieve full remote code execution (RCE).
Affected Software
Event History
Frequently Asked Questions
Who is exposed to this issue?
DocsGPT deployments running version 0.15.0 or below are exposed if the custom prompt feature is available to unauthenticated users. No authentication is required for exploitation.
What does an attacker need to exploit it?
An attacker needs the ability to submit malicious content through the custom prompt feature. The injected content is rendered as a Jinja template without sanitization or sandboxing.
What is the likely impact of successful exploitation?
Successful template injection can lead to server-side template injection and full remote code execution on the affected server.