Broken Access Control in FastGPT AI Agent Platform
CVE-2026-40252

5.3MEDIUM

Key Information:

Vendor

Labring

Status
Fastgpt
Vendor
CVE Published:
10 April 2026

What is CVE-2026-40252?

FastGPT, a platform designed for building AI agents, suffers from a Broken Access Control vulnerability prior to version 4.14.10.4. This flaw allows authenticated users to access and execute applications that belong to other teams simply by providing an incorrect application ID (appId). Although the API does validate the team token, it fails to ensure that the requested application matches the authenticated team, resulting in potential cross-tenant data exposure and unauthorized execution of sensitive AI workflows.

Affected Version(s)

FastGPT < 4.14.10.4

References

https://github.com/labring/FastGPT/security/advisories/GH...
x_refsource_CONFIRM
https://github.com/labring/FastGPT/releases/tag/v4.14.10.4
x_refsource_MISC

CVSS V4

Score:
5.3
Severity:
MEDIUM
Confidentiality:
Low
Integrity:
None
Availability:
Low
Attack Vector:
Network
Attack Complexity:
Low
Attack Required:
None
Privileges Required:
Undefined
User Interaction:
None

Timeline

  • Vulnerability published

  • Vulnerability Reserved

.