CVE-2026-31951: LibreChat's MCP Server Header Injection Enables OAuth Token Theft
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 {{LIBRECHATOPENIDACCESSTOKEN}} (and others), causing victims who call tools on that server to have their OAuth tokens exfiltrated. Version 0.8.3-rc2 fixes the issue.
Affected Software
Event History
Frequently Asked Questions
What is the severity of CVE-2026-31951?
CVE-2026-31951 is classified as a medium severity vulnerability due to its potential for OAuth token theft.
How do I fix CVE-2026-31951?
To fix CVE-2026-31951, update LibreChat to the latest version that addresses this vulnerability.
What versions of LibreChat are affected by CVE-2026-31951?
CVE-2026-31951 affects LibreChat versions 0.8.2-rc1 through 0.8.3-rc1.
What is the impact of CVE-2026-31951 on user data?
CVE-2026-31951 can lead to unauthorized access to user OAuth tokens, compromising user authentication.
Is there a workaround for CVE-2026-31951 if I cannot update?
Currently, there is no documented workaround for CVE-2026-31951, so it's recommended to apply the update as soon as possible.