GHSA-mhr3-j7m5-c7c9: LangGraph: BaseCache Deserialization of Untrusted Data may lead to Remote Code Execution

Summary

LangGraph's caching layer can deserialize attacker-controlled cache entries with pickle.loads when msgpack serialization fails. Before langgraph-checkpoint 4.0.0, BaseCache defaults to JsonPlusSerializer(pickle_fallback=True), so exploitation leads to remote code execution in the LangGraph process. Exploitation requires that caching is enabled and that the attacker can write to the cache backend, making it a post-compromise escalation vector.

Mitigation

Upgrade to langgraph-checkpoint>=4.0.0.

Affected packages in the Exposure Registry

Matched by package name and ecosystem. Each entry shows whether the package delegates to a language model and how many tracked packages depend on it.

• langgraph-checkpointPyPILLM dependency since 2024-08-02 · 7 tracked dependents

Related items

• MediumGHSA-8prr-f582-jqq3: LangGraph CLI prints the registry push token in `langgraph deploy --verbose` outputSame vendor · GitHub Advisory Database
• HighGHSA-4hm6-w6qq-w73v: LangGraph JavaScript custom authentication may permit cross-user Store accessSame vendor · GitHub Advisory Database
• MediumGHSA-3fx2-cqw9-xr3c: LangGraph: Unencoded identifiers in LangGraph SDK v3 streaming can redirect authenticated GET requestsSame vendor · GitHub Advisory Database
• MediumGHSA-6hmw-9432-67qq: LangGraph: Namespace labels containing `.` reach other namespaces in Postgres and SQLite storesSame vendor · GitHub Advisory Database
• MediumCVE-2026-108575: BerriAI LiteLLM improper authorization in secret resolution via api_keySame vendor · NVD/CVE Database