Cyber Resilience

CVE-2026-31944

HighPublic PoC

Published: 13 March 2026

Published
13 March 2026
Modified
17 March 2026
KEV Added
Patch
CVSS Score v3.1 7.6 CVSS:3.1/AV:N/AC:L/PR:L/UI:R/S:C/C:H/I:L/A:N
EPSS Score 0.0004 12.6th percentile
Risk Priority 15 60% EPSS · 20% KEV · 20% CVSS

Summary

CVE-2026-31944 is a high-severity Missing Authentication for Critical Function (CWE-306) vulnerability in Librechat Librechat. Its CVSS base score is 7.6 (High).

Operationally, exploitation aligns with the MITRE ATT&CK technique Exploit Public-Facing Application (T1190); ranked at the 12.6th percentile by exploit likelihood (below the median); it is not currently listed in the CISA KEV catalog; a public proof-of-concept is referenced.

This vulnerability is AI-related — categorised as LLM Application Platforms; in the Protocol-Specific Risks risk domain.

The strongest mitigations our analysis identified are NIST 800-53 AC-14 (Permitted Actions Without Identification or Authentication) and AC-3 (Access Enforcement).

Deeper analysis

CVE-2026-31944 is a missing authentication vulnerability in the MCP (Model Context Protocol) OAuth callback endpoint of LibreChat, an open-source ChatGPT clone with additional features. Versions from 0.8.2 to 0.8.2-rc3 are affected, as the endpoint accepts redirects from identity providers and stores OAuth tokens for the user who initiated the flow without verifying that the browser is logged in or that the logged-in user matches the initiator. This flaw, classified under CWE-306 (Missing Authentication for Critical Function), carries a CVSS v3.1 base score of 7.6 (AV:N/AC:L/PR:L/UI:R/S:C/C:H/I:L/A:N).

A low-privileged LibreChat account holder can exploit the vulnerability by sending a malicious authorization URL to a victim user. When the victim completes the OAuth flow via their browser, the resulting tokens are stored on the attacker's LibreChat account rather than the victim's, granting the attacker access to the victim's MCP-linked services, such as Atlassian or Outlook, and enabling account takeover.

Published on 2026-03-13, the issue was addressed in LibreChat version 0.8.3-rc1. The official GitHub security advisory at https://github.com/danny-avila/LibreChat/security/advisories/GHSA-vf7j-7mrx-hp7g provides details on the patch and recommends upgrading to the fixed release for mitigation.

EU & UK References

Vulnerability details

LibreChat is a ChatGPT clone with additional features. From 0.8.2 to 0.8.2-rc3, The MCP (Model Context Protocol) OAuth callback endpoint accepts the redirect from the identity provider and stores OAuth tokens for the user who initiated the flow, without verifying…

more

that the browser hitting the redirect URL is logged in or that the logged-in user matches the initiator. An attacker can send the authorization URL to a victim; when the victim completes the flow, the victim’s OAuth tokens are stored on the attacker’s LibreChat account, enabling account takeover of the victim’s MCP-linked services (e.g. Atlassian, Outlook). This vulnerability is fixed in 0.8.3-rc1.

CWE(s)

AI Security AnalysisAI

AI Category
LLM Application Platforms
Risk Domain
Protocol-Specific Risks
OWASP Top 10 for LLMs 2025
None mapped
Classification Reason
Matched keywords: chatgpt, librechat, mcp, model context protocol

Related Threats

MITRE ATT&CK Enterprise TechniquesAI

T1190 Exploit Public-Facing Application Initial Access
Adversaries may attempt to exploit a weakness in an Internet-facing host or system to initially access a network.
T1528 Steal Application Access Token Credential Access
Adversaries can steal application access tokens as a means of acquiring credentials to access remote systems and resources.
T1550.001 Application Access Token Lateral Movement
Adversaries may use stolen application access tokens to bypass the typical authentication process and access restricted accounts, information, or services on remote systems.
Why these techniques?

Missing auth on OAuth callback enables remote exploitation of public-facing app (T1190) to misdirect and steal application access tokens (T1528) for use as alternate auth material on linked services (T1550.001).

Confidence: MEDIUM · MITRE ATT&CK Enterprise v18.1

CVEs Like This One

CVE-2026-22252Same product: Librechat Librechat
CVE-2026-31945Same product: Librechat Librechat
CVE-2025-69220Same product: Librechat Librechat
CVE-2025-41258Same product: Librechat Librechat
CVE-2025-69222Same product: Librechat Librechat
CVE-2026-4276Same product: Librechat Librechat
CVE-2024-10361Same product: Librechat Librechat
CVE-2026-33265Same product: Librechat Librechat
CVE-2026-31943Same product: Librechat Librechat
CVE-2026-31942Same product: Librechat Librechat

Affected Assets

librechat
librechat
0.8.2

Mitigating Controls

Mitigating Controls (NIST 800-53 r5) AI

prevent

Enforces approved authorizations by requiring verification that the user is logged in and matches the OAuth flow initiator before storing tokens on the MCP callback endpoint.

prevent

Ensures communications session authenticity by binding the OAuth callback session identifier to the initiator's user identity, preventing token storage mismatches.

prevent

Limits permitted actions without identification or authentication, prohibiting unauthenticated access to the critical MCP OAuth callback endpoint.

References