NewApp Monitoring is liveSign up free
AgentbeamBeta
Open-source package vulnerability · npm

CVE-2026-106119: LangChain: MongoDBChatMessageHistory query injection can allow cross-session access

mediumCVSS 4.0 6.0GHSA-m6rx-h84q-8r95

Impact MongoDBChatMessageHistory did not enforce the documented string type for session identifiers at runtime. In affected applications, a structured session identifier could be interpreted as a MongoDB query condition rather than as a literal identifier.

Published Oct 8, 2026

Affected packages

@langchain/mongodb

Ecosystem
npm
Affected
<= 1.3.0
Fixed in
1.3.1
Check
npm ls @langchain/mongodb

Advisory details

From the GitHub Advisory Database (CC BY 4.0).

Impact

MongoDBChatMessageHistory did not enforce the documented string type for session identifiers at runtime. In affected applications, a structured session identifier could be interpreted as a MongoDB query condition rather than as a literal identifier.

Applications are affected when they pass untrusted session identifier input to MongoDBChatMessageHistory and store multiple users’ histories in the same collection. An attacker who can invoke chat-history operations may be able to read, modify, or delete another user’s stored conversation.

Applications that derive the session identifier from an authenticated, server-controlled value and enforce its type are not affected by this path.

Patches

The issue is fixed in @langchain/mongodb version 1.3.1.

Users of affected versions should upgrade to version 1.3.1 or later. The patched version validates session identifiers at runtime and ensures they are compared as literal values in MongoDB queries.

Workarounds

If an immediate upgrade is not possible, applications must reject session identifiers that are not non-empty strings. Session identifiers should be derived from an authenticated, server-controlled value and must not be passed directly from an untrusted request field.

Upgrading remains the recommended remediation.

Credited to iaohkut-from-NightWolf-Team.

References

How this page is made: the data panel comes from the GitHub Advisory Database, CISA’s KEV catalog, the NVD and FIRST EPSS, and updates twice a day. No analysis has been written for this entry yet. Methodology