CVE-2026-31944
Received
Received - Intake
OAuth Token Theft via MCP Callback in LibreChat
Publication date: 2026-03-13
Last updated on: 2026-03-17
Assigner: GitHub, Inc.
Description
Description
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 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 victimβs OAuth tokens are stored on the attackerβs LibreChat account, enabling account takeover of the victimβs MCP-linked services (e.g. Atlassian, Outlook). This vulnerability is fixed in 0.8.3-rc1.
CVSS Scores
EPSS Scores
| Probability: | |
| Percentile: |
Meta Information
Affected Vendors & Products
| Vendor | Product | Version / Range |
|---|---|---|
| librechat | librechat | 0.8.2 |
| librechat | librechat | 0.8.2 |
| librechat | librechat | 0.8.2 |
| librechat | librechat | 0.8.2 |
Helpful Resources
Exploitability
| CWE ID | Description |
|---|---|
| CWE-306 | The product does not perform any authentication for functionality that requires a provable user identity or consumes a significant amount of resources. |