CVE-2026-31951: medium-severity vulnerability in danny-avila LibreChat
LibreChat's MCP Server Header Injection Enables OAuth Token Theft
Published · Updated
E-mail alert when it lands in KEV, gets a public PoC or its score goes up.
13Vexday Risk Score
No sign of exploitation. No public exploitation artifact known so far.
ssvc Trackcvss 6.8epss 0.3%
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.
CVSS:3.1/AV:N/AC:L/PR:L/UI:R/S:C/C:H/I:N/A:N
Affected products
danny-avila · LibreChatRelated CVEs — danny-avila LibreChat
In the same product, most dangerous first.
CVE-2025-69222CRITICALLibreChat is vulnerable to Server-Side Request Forgery due to missing restrictionsEPSS 4.6%CVE-2026-22252CRITICALLibreChat MCP Stdio Remote Command ExecutionEPSS 4.2%CVE-2026-32625CRITICALLibreChat Exfiltrates Server Secrets via MCP Server URL InjectionEPSS 2.8%CVE-2025-54868HIGHLibreChat exposes arbitrary chats through Meilisearch engineEPSS 0.5%CVE-2026-54037MEDIUMLibreChat: Incomplete Fix for CVE-2025-7105 — /api/convos/duplicate Lacks Rate Limiting Applied to /api/convos/forkEPSS 0.5%CVE-2026-54024MEDIUMLibreChat: Incomplete Fix for CVE-2024-11171 — Conversation Import Multer Instance Missing File Size LimitsEPSS 0.4%