CVE Catalog

CVE-2026-55253

HighCVSS 7.7
Published: Updated: Translated: NVD NIST

Exploitation Probability (EPSS)

Low risk
0.53%

43th percentile - higher than 43% of all known CVEs

Summary

In LangChain MongoDB integrations prior to langgraph-checkpoint-mongodb 0.3.0 and langgraph-store-mongodb 0.4.0, MongoDBSaver.list(), alist(), and MongoDBStore.search() incorporate filter dictionaries into MongoDB queries without rejecting keys prefixed with $. An authenticated caller controlling a filter can inject MongoDB operators such as $regex or $where, bypassing tenant isolation.

Risk Assessment

In a multi-tenant deployment, an attacker can bypass equality filters and access other tenants' data, causing a serious confidentiality breach.

Recommendation

Upgrade langgraph-checkpoint-mongodb to 0.3.0 and langgraph-store-mongodb to 0.4.0, and until then validate user-supplied filters for keys prefixed with $.

Related vulnerabilities

Original NVD description (English source)

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.

Vulnerability data from NVD (NIST) · CISA KEV · EPSS