Overview
The vulnerability in focus, CVE-2025-54868, targets LibreChat, a popular ChatGPT clone. The issue lies within an unprotected testing endpoint that could potentially expose the chats of arbitrary users to malicious parties. Given the widespread use of LibreChat, this vulnerability can have grave consequences, warranting immediate attention and remediation.
Vulnerability Summary
CVE ID: CVE-2025-54868
Severity: High (7.5 CVSS Score)
Attack Vector: Network
Privileges Required: None
User Interaction: None
Impact: Unauthorized access to user chats could lead to system compromise and data leakage
Affected Products
Escape the Surveillance Era
Most apps won’t tell you the truth.
They’re part of the problem.
Phone numbers. Emails. Profiles. Logs.
It’s all fuel for surveillance.
Ameeba Chat gives you a way out.
- • No phone number
- • No email
- • No personal info
- • Anonymous aliases
- • End-to-end encrypted
Chat without a trace.
Product | Affected Versions
LibreChat | 0.0.6 to 0.7.7-rc1
How the Exploit Works
The exploit takes advantage of the endpoint /api/search/test in LibreChat, which allows direct access to stored chats in the Meilisearch engine without the requirement of proper access control. By sending a crafted HTTP request to the exposed endpoint, an attacker can read arbitrary chats from the system, thereby breaching user privacy and potentially gaining sensitive information.
Conceptual Example Code
Below is a conceptual HTTP request an attacker might use to exploit this vulnerability:
GET /api/search/test?query=<user_id> HTTP/1.1
Host: target.librechat.com
In this example, `

