Authorization Flaw in LangGraph API Affects Langchain Products
CVE-2026-55236

5.9MEDIUM

Key Information:

Vendor

Langchain-ai

Status
Langgraph-api
Vendor
CVE Published:
14 September 2026

What is CVE-2026-55236?

The LangGraph API before version 0.10.0 contains an authorization vulnerability in the run-creation path, which allows low-privileged users to access another user's private assistant. This occurs due to the erroneous use of assistants.search instead of the correct assistants.read, leading to insufficient access control checks. As a result, sensitive information such as metadata, configuration, and context of private assistants can be exposed during the execution of runs. Deployments that do not have custom authorization handlers or global owner filters in place are particularly vulnerable. This issue has been addressed in version 0.10.0.

Affected Version(s)

langgraph-api < 0.10.0

References

https://github.com/langchain-ai/helm/security/advisories/...
x_refsource_CONFIRM
https://github.com/langchain-ai/docs/blob/7b70fbacab21f84...
x_refsource_MISC

CVSS V3.1

Score:
5.9
Severity:
MEDIUM
Confidentiality:
High
Integrity:
Low
Availability:
High
Attack Vector:
Network
Attack Complexity:
High
Privileges Required:
Low
User Interaction:
None
Scope:
Unchanged

Timeline

  • Vulnerability published

  • Vulnerability Reserved

.