Skip to content
OpenAdvisories
High · 7.7 CVE-2026-55253 EUVD-2026-77674GHSA-533j-2v4q-mw5hPYSEC-2026-3859PYSEC-2026-3860

LangChain MongoDB: NoSQL Operator Injection in MongoDBSaver.list() leading to cross-tenant data exposure

Published Sep 14, 2026 · Updated Sep 14, 2026 · Assigned by GitHub_M

Description

LangChain MongoDB provides integrations between MongoDB, Atlas, LangChain, and LangGraph. Prior to langgraph-checkpoint-mongodb 0.3.0 and langgraph-store-mongodb 0.4.0, MongoDBSaver.list(), MongoDBSaver.alist(), and MongoDBStore.search() incorporate filter dictionaries into MongoDB queries without recursively rejecting keys prefixed with $. An authenticated caller who controls a filter argument through HTTP query parameters, request body fields, or agent tool arguments can inject MongoDB Query Language operators such as $regex or $where. In a multi-tenant deployment that uses the filter to enforce per-user or per-tenant isolation, injected operators can bypass intended equality filtering and expose other tenants' checkpoint or store data. Filters constructed entirely from trusted server-side values have lower practical risk. This issue is fixed in langgraph-checkpoint-mongodb 0.3.0 and langgraph-store-mongodb 0.4.0.

Affected products

VendorProductAffectedFixed
langchain-ai langchain-mongodb < 0.4.0 —
langchain-ai langgraph-checkpoint-mongodb < 0.3.0 —
langchain-ai langgraph-store-mongodb < 0.4.0 —

Affected packages

EcosystemPackageAffectedFixedRecord
PyPI langgraph-checkpoint-mongodb < 0.3.0 0.3.0 GHSA-533j-2v4q-mw5h
PyPI langgraph-store-mongodb < 0.4.0 0.4.0 GHSA-533j-2v4q-mw5h

References