Skip to content
WatchlistHigh

CVE-2025-68664langchain Insecure Deserialization

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

Published Dec 23, 2025 · Updated Sep 10, 2026 · Also known as GHSA-c67j-w6g6-q2cm, PYSEC-2026-373, ROOT-APP-PYPI-CVE-2025-68664

Key Facts

Patch immediately
  • CVE-2025-68664 is a high-severity vulnerability (CVSS 8.2 v3.1) affecting Langchain Core, py3-langchain (chainguard), ansible-automation-platform (redhat) and 5 more products.
  • It is not listed in any known-exploited vulnerabilities catalog. Public exploit code is available.
  • EPSS estimates a 43% probability of exploitation in the next 30 days.
  • Fixed versions are available, including ansible-automation-platform 26/lightspeed-rhel9:1767625654, Langchain Core 0.3.81, Langchain Core 0.1.23+sp999.
  • Brandefense priority: patch immediately, because public exploit code exists for a high-severity flaw.

Scores

CVSS v3.1

8.2

CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:L/A:N

EPSS (30-day exploit likelihood)

43%

Higher than 98.7% of all scored CVEs

Exploit maturity

Proof of concept

ATLAS score

45/100

Publication score, see How it works

Exploitation Verdict

KEV listed

Not listed

Federal due date

None

No federal due date

Ransomware use

Not reported

Attributed actors

None named

Exploit & Defence Guidance

Exploit maturity

Proof of concept

Public exploit code

1

Proof-of-concept links

In-the-wild evidence

0

Reports of active use

Description

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 Products

12 affected entries in total
ProductAffected versionsFixed in
Langchain CoreLangchain>= 1.0.00.3.81
py3-langchainchainguardSee advisory1.2.6-r0
ansible-automation-platformredhatSee advisory26/lightspeed-rhel9:1767625654
openshift-lightspeed/lightspeed-service-api-rhel9:1769058587redhatSee advisoryn/a
openshift-lightspeed/lightspeed-service-api-rhel9:sha256:23af5c023d054ad9f607f4a8df270c01607a7ace1acd328b5c3702e32dccb621redhatSee advisoryn/a
3 more affected products in the full CVE brief

Patched Releases

ansible-automation-platform

26/lightspeed-rhel9:1767625654

Langchain Core

0.3.81, 0.1.23+sp999, 0.3.72+aikido.2, 0.3.72+aikido.3, 0.3.72+aikido.4, 0.3.78+aikido.2, 1.0.2+aikido.4, 1.0.2+aikido.5, 1.2.0+aikido.2, 1.2.4+aikido.5, 1.2.5

py3-langchain

1.2.6-r0

rootio-langchain-core

0.3.72+root.io.1, 0.3.72+root.io.2, 0.3.78+root.io.1, 1.0.2+root.io.1, 1.0.2+root.io.3, 1.0.2+root.io.4, 1.2.0+root.io.1, 1.2.4+root.io.1, 1.2.4+root.io.2, 1.2.4+root.io.3, 1.2.4+root.io.4, 1.2.4+root.io.5

seal-langchain-core

0.1.23+sp999

Timeline

  1. Publicly disclosed

    Dec 23, 2025

  2. Last updated

    Sep 10, 2026

Weakness Type

  • CWE-502 Deserialization of Untrusted Data

References

12 more references in the full CVE brief

FAQ

Common Questions.

Is CVE-2025-68664 On Your Attack Surface?

Brandefense matches it against the products discovered on your perimeter and opens a finding the day a CVE like this is listed.

Request A Demo

Figures as of Oct 19, 2026. Compiled by Brandefense technologies from public vulnerability catalogs; scoring as described in How it works.

To report an error, contact Brandefense.