Arbitrary Header Injection in LibreChat by vendor Danny Avila
CVE-2026-31951

6.8MEDIUM

Key Information:

Vendor

Danny-avila

Status
Librechat
Vendor
CVE Published:
27 March 2026

What is CVE-2026-31951?

The vulnerability in LibreChat allows attackers to create malicious Model Context Protocol (MCP) servers that can inject arbitrary HTTP headers. This flaw affects versions 0.8.2-rc1 through 0.8.3-rc1, enabling exfiltration of sensitive OAuth tokens, such as {{LIBRECHAT_OPENID_ACCESS_TOKEN}}, when users interact with compromised servers. The vulnerability has been addressed in version 0.8.3-rc2.

Affected Version(s)

LibreChat >= v0.8.2-rc1, <= v0.8.3-rc1

References

https://github.com/danny-avila/LibreChat/security/advisor...
x_refsource_CONFIRM

CVSS V3.1

Score:
6.8
Severity:
MEDIUM
Confidentiality:
High
Integrity:
None
Availability:
High
Attack Vector:
Network
Attack Complexity:
Low
Privileges Required:
Low
User Interaction:
Required
Scope:
Changed

Timeline

  • Vulnerability published

  • Vulnerability Reserved

.