CVE-2026-31951 MEDIUM

CVE-2026-31951: LibreChat's MCP Server Header Injection Enables OAuth Token Theft

Vendor Danny-Avila
Product LibreChat
Weakness CWE-200 · Info exposure
Published March 27, 2026
Last update March 31, 2026
View on NVD All CVEs

CVSS base score

6.8/10
Attack vector Network
Attack complexity Low
Privileges required Low
User interaction Required
Confidentiality High
Integrity None

CVSS vector

CVSS:3.1/AV:N/AC:L/PR:L/UI:R/S:C/C:H/I:N/A:N

What the vulnerability does

01Description

LibreChat is a ChatGPT clone with additional features. In versions 0.8.2-rc1 through 0.8.3-rc1, user-created MCP (Model Context Protocol) servers can include arbitrary HTTP headers that undergo credential placeholder substitution. An attacker can create a malicious MCP server with headers containing `{{LIBRECHAT_OPENID_ACCESS_TOKEN}}` (and others), causing victims who call tools on that server to have their OAuth tokens exfiltrated. Version 0.8.3-rc2 fixes the issue.

Key dates

02Disclosure timeline

March 27, 2026 CVE published
March 31, 2026 Record updated