Executive brief
LibreChat, an open-source interface for multiple AI providers, contains a flaw that allows users to upload files of any size when importing conversations. An authenticated user can exploit this to upload massive files that fill up the server's storage or crash its memory. This can lead to a complete service outage, preventing all users from accessing the AI chat platform.
Technical details
The vulnerability exists because the 'POST /api/convos/import' endpoint uses a 'multer' instance that lacks the 'limits: { fileSize }' configuration, which was previously applied to other upload routes in a partial fix for CVE-2024-11171. Furthermore, the application-level size check in 'importConversations.js' is ineffective because the 'CONVERSATION_IMPORT_MAX_FILE_SIZE_BYTES' environment variable is disabled by default, causing the check to evaluate to false. Even if enabled, the check occurs after the file has already been written to disk and loaded into memory via 'fs.readFile()'. An authenticated attacker can trigger disk exhaustion, Out-of-Memory (OOM) crashes, or CPU exhaustion by uploading multi-gigabyte JSON files. The issue is resolved in version 0.8.4-rc1.
Affected products
- danny-avila LibreChat < 0.8.4-rc1
Timeline
- 2026-06-02: advisory: GitHub Security Advisory published
- 2026-06-25: disclosed: CVE published to NVD