CVE-2025-2334 is a medium severity (CVSS 5.4) improper access control flaw in the deleteChat function of the springboot-openai-chatgpt project. The chatListId argument is not properly checked, and the record lists no fixed version. If you run this code, add the authorization check yourself or retire the project.
What's affected
The record, written by VulDB, says the issue affects the function deleteChat, reached at /api/mjkj-chat/chat/ai/delete/chat in the Chat History Handler component, in springboot-openai-chatgpt at commit e84f6f5. Manipulating the chatListId argument leads to improper access control. The attack can be started remotely and needs a low privilege account.
Unlike its sibling CVE-2025-2322, this record does not state that the product lacks versioning or that the vendor is silent in its text, so check the project's repository for any later commit. The VulDB page could not be loaded for this review, and no fixed version appears in the record.
How to check if you're exposed
Review the code for the delete handler. You are looking for one question: does the code check that the chat identified by chatListId belongs to the signed in user before deleting it?
grep -rn "deleteChat" src/
grep -rn "chatListId" src/
In a running instance, test with two ordinary test accounts in a non production environment. If one account can delete the other's chat by changing the identifier, you have the issue.
Reading an inventory is faster than checking host by host. Safeguard has a free in-browser dependency scan at gold.safeguard.sh/scan and a Gold directory you can search for any CVE. Reachability analysis helps you see whether a vulnerable component is actually used in your code, so you can fix the exposed ones first.
What to do now
- Locate every deployment of the project and its forks.
- Add an ownership check to the delete handler so a user can only remove their own chats.
- Apply the same review to the other handlers that accept an identifier from the client.
- Log delete operations with the user and identifier so misuse is visible.
- Plan to replace the project with a maintained one if you depend on it.
Why it matters
CWE-284 is improper access control and CWE-266 is incorrect privilege assignment. The pattern is common: the server trusts an identifier sent by the client and acts on it without checking who owns the object.
The impact is limited, which is why the score is 5.4. A user could alter or remove chat history that is not theirs. In an application that stores conversations with an AI model, history can contain sensitive prompts, so the integrity of that data matters. The KEV flag is false and no exploitation is claimed.
Quick check
Does a fixed version exist?
The record lists none. Check the project's repository and add the check yourself if you run the code.
What kind of bug is this?
Improper access control. The delete handler acts on an identifier without checking ownership.
Sources
Last reviewed: 19 October 2026.
