CVE-2025-68664

What it is

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.

At a glance

Exploitability

Can be exploited over the network without logging in. High risk of data exposure.

Exploit odds

42.9% chance of exploitation in the next 30 days (EPSS). Higher than most.

Fix

A patch, advisory or fixed version is referenced. Upgrade or apply it.

Timeline

1. 2025-12-22 23:28 UTC ReservedThe CVE id was reserved by its CNA
2. 2025-12-23 23:15 UTC PublishedDisclosed publicly on CVE.org and NVD
3. 2026-10-19 22:10 UTC Last updatedMost recent change to the record

Details

Published

2025-12-23 23:15 UTC

Last modified

2026-10-19 22:10 UTC

Exploit probability

42.9% · EPSS, next 30 days

Weakness

CWE-502 · Unsafe deserialization

Record status

Analyzed

Affected

• Langchain Core < 0.3.81
• Langchain Core < 1.2.5

Attack profile CVSS 3.1 · 9.3

Attack vector

Network

Complexity

Low

Privileges needed

None

User interaction

None

Scope

Changed

Confidentiality

High

Integrity

Low

Availability

None