Machine-written summary
Summary written by Claude Haiku from the source advisory. Not reviewed by a person.
The langchain-core package, versions before 0.3.85 and from 1.0.0 up to 1.3.3, has overly broad object allowlists in some runtime code paths that deserialize run data. An attacker who can submit untrusted structured input that gets preserved in run inputs or outputs, in an application using RunnableWithMessageHistory, astream_log(), or astream_events(version="v1"), may be able to get trusted LangChain classes instantiated with attacker-controlled arguments. Upgrade langchain-core to 0.3.85 or 1.3.3.
Not analysed yet
We haven’t published our analysis of this Advisory yet. Once a person has reviewed it, this page will say what’s vulnerable, what has to be true for your app to be affected, and how to check.
Severity and exploitation
| Score | Rating | CVSS | Vector | Source |
|---|---|---|---|---|
| 8.2 | High | 3.1 | CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:L/A:N | CNA (GitHub_M) |
- CISA KEV Not listed.
- EPSS 0.0041, the 35th percentile, on 11 Nov 2026: FIRST’s estimate of the chance of exploitation in the next 30 days.
Timeline
- 5 May 2026, 21:00 UTC Published by the maintainers as a GitHub repository advisory
- 31 Oct 2026, 14:11 UTC First seen by agentprime
- 31 Oct 2026, 14:11 UTC Confirmed by OSV