Vulnerability Assessment
| Weakness |
CWE
OWASP Top 10:2025
A01:2025 - Broken Access Control
ASVS 5.0
V8 - Authorization Access control enforcement
MITRE ATT&CK
1 technique across 1 tactic: Defense Evasion The technique list with ATT&CK links and the kill-chain view is part of the Basic plan and up. |
| Impact | The CVSS score of 8.0 (High) reflects the vulnerability's significant impact with adjacent network attack vector (AV:A), low complexity (AC:L), and requiring only low privileges (PR:L). The attack enables high confidentiality, integrity, and availability impact (C:H/I:H/A:H) without user interaction. However, real-world risk is tempered by several factors: the attack requires adjacent network access and authenticated user credentials, limiting the attack surface compared to remote unauthenticated exploits. The vulnerability affects a specific release candidate version (0.8.1-rc2) rather than a stable production release, potentially reducing the number of deployed instances. While the technical impact is severe for affected installations allowing lateral movement from user sessions to RAG API access, the adjacent network requirement and authentication prerequisite make this less immediately critical than remotely exploitable unauthenticated vulnerabilities, though it remains a serious concern for multi-tenant or shared hosting environments. |
| Exploit ScenarioDerived | An authenticated user with low-privilege access to LibreChat on an adjacent network obtains their own session JWT token through normal authentication. By analyzing the token structure and reusing the same JWT secret that signs both user sessions and RAG API requests, the attacker crafts or repurposes their session token to authenticate directly to the RAG API endpoints, bypassing service-level access controls. This allows the attacker to read sensitive RAG data, modify retrieval configurations, or disrupt the RAG service availability, potentially accessing other users' context data or poisoning the knowledge base used for AI-augmented responses across the application. |
Get an e-mail when this CVE is exploited, patched or re-scored. Free account.
Alert meAnalysisAIv1 – Mar 18, 2026
A critical authentication bypass vulnerability exists in LibreChat version 0.8.1-rc2 where the same JWT secret is reused for both user session management and the RAG (Retrieval-Augmented Generation) API authentication. This design flaw allows authenticated users to compromise service-level authentication of the RAG API by leveraging their session tokens to access or manipulate the RAG service beyond intended privileges. No active exploitation (KEV) has been reported, but a detailed security advisory with technical analysis is publicly available from SBA Research.
RemediationAI
Users running LibreChat version 0.8.1-rc2 should immediately upgrade to a patched version or implement separate JWT secrets for user session management and RAG API authentication by modifying the application configuration to use distinct cryptographic keys for each authentication domain. Review the SBA Research security advisory at https://github.com/sbaresearch/advisories/tree/public/2025/SBA-ADV-20251205-01_LibreChat_RAG_API_Authentication_Bypass for detailed technical guidance and consult the official LibreChat repository at https://github.com/danny-avila/LibreChat for patch availability and upgrade instructions. As an interim mitigation, restrict network access to the RAG API endpoints to only trusted service accounts and implement additional authorization checks to validate that JWT tokens used for RAG API access originate from legitimate service contexts rather than user sessions.
Affected VersionsENISA
Affected Packages
Source: OSV (Open Source Vulnerabilities)
Affects your stack
Affects 1 package in euvd
Checking this CVE against your own stacks and servers is part of the Basic plan and up.
Technical ContextAI · show details
LibreChat is a web-based chat application (cpe:2.3:a:danny-avila:librechat) that implements RAG (Retrieval-Augmented Generation) functionality for enhanced AI responses. The vulnerability stems from CWE-284 (Improper Access Control) where the application uses a shared JWT signing secret across different security boundaries. When the same cryptographic key material is used to sign tokens for user sessions and service-level API authentication, users can potentially forge or reuse their session JWTs to authenticate to backend services that should have separate authentication mechanisms. This violates the principle of defense in depth and breaks service isolation.
Know when Librechat gets its next vulnerability
- Watch products and get an email when something new is published
- Check this CVE against your own stack and servers
- Exploit and PoC references, threat intelligence from 22 sources
No credit card · 7-day full trial
Severity by source
AV:A/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H
Primary rating from NVD · only source for this CVE.
CVSS VectorNVD
DescriptionCVE.org
LibreChat version 0.8.1-rc2 uses the same JWT secret for the user session mechanism and RAG API which compromises the service-level authentication of the RAG API.
Executive ViewAI
LibreChat version 0.8.1-rc2 contains a critical authentication bypass that allows low-privilege users to gain unauthorized administrative access to RAG API functionality, potentially exposing sensitive data and enabling service manipulation. Any organization running this version with internet or adjacent network exposure faces immediate risk of data breach and system compromise.
Implement strict network segmentation restricting RAG API access to trusted internal networks only; deploy Web Application Firewall rules to enforce separate authentication mechanisms or tokens for RAG API distinct from session JWTs; disable RAG functionality entirely if business operations permit; implement enhanced API monitoring and alerting for RAG endpoint access patterns; require multi-factor authentication for administrative functions; rotate all JWT secrets immediately and monitor for unauthorized token generation.
Threat Intelligence Overview
How this decision was reached
The engine reads each signal below and applies a fixed list of rules from top to bottom. The first rule that matches sets the recommended action – like a statement, every line is shown.
What every action level means
SSVC AssessmentENISA
Priority Score
Attack Technique
Authentication bypass attacks exploit flaws in the verification mechanisms that control access to systems and applications.
Related Vulnerabilities in LibreChat JWT and Authorization Flaws
View all 2A vulnerability in danny-avila/librechat version git 81f2936 allows for path traversal due to improper sanitization of f
An arbitrary file deletion vulnerability exists in danny-avila/librechat version v0.7.5-rc2, specifically within the /ap
LibreChat before v0.8.2-rc2 allows any authenticated user to execute shell commands as root inside the container through
LibreChat 0.7.9 lets an authenticated user turn off two-factor authentication on their own account without supplying a v
Stored cross-site scripting and tracker injection in LibreChat 0.8.0 and earlier allow a user who can create a chat to c
LibreChat is a ChatGPT clone with additional features. Rated high severity (CVSS 8.6), this vulnerability is remotely ex
LibreChat 0.8.1-rc2 has SSRF in the Actions feature that allows authenticated users to make the server perform requests
In danny-avila/librechat version git 0c2a583, there is an improper input validation vulnerability. Rated high severity (
A mass assignment vulnerability exists in danny-avila/librechat, affecting all versions. Rated high severity (CVSS 7.5),
Unbounded `key` and `value` parameters on the LibreChat 0.7.9 memory API (`POST /api/memories` and `PATCH /api/memories/
An unhandled exception in the danny-avila/librechat repository, version git 600d217, can cause the server to crash, lead
An improper access control vulnerability (IDOR) exists in the delete attachments functionality of danny-avila/librechat
EUVD-2025-208825
NVD
CVE.org
EPSS
GitHub
VulDB
Exploit-DB