CVE-2025-68664

published 23 Dec 2025 · updated 24 Dec 2025

LangChain serialization injection vulnerability enables secret extraction in dumps/loads APIs

10.0CriticalCVSS 9.3EPSS 43%2 public exploits

2 public exploits exists; no exploitation in the wild has been reported. EPSS puts the chance of exploitation in the next 30 days at 43%. No fixed version is recorded.

LangChain is a framework for building agents and LLM-powered applications. Prior to versions 0.3.81 and 1.2.5, a serialization injection vulnerability exists in LangChain's dumps() and dumpd() functions. The functions do not escape dictionaries with 'lc' keys when serializing free-form dictionaries. The 'lc' key is used internally by LangChain to mark serialized objects. When user-controlled data contains this key structure, it is treated as a legitimate LangChain object during deserialization rather than plain user data. This issue has been patched in versions 0.3.81 and 1.2.5.

Affected

Vendors, products and packages with version ranges

VendorProductVersionsFixed
langchain-ai<0.3.81, >=1.0.0 <1.2.5

Exploit evidence

2 public sources

Timeline

Every event on this vulnerability

Embed

A live severity badge for an advisory, README or ticket

CVE-2025-68664 severity on Zeptric[![CVE-2025-68664](https://zeptric.io/badge/cve/CVE-2025-68664.svg)](https://zeptric.io/cve/CVE-2025-68664)