Code Execution Vulnerability in FastGPT's Sandbox Environment
CVE-2025-49131

6.3MEDIUM

Key Information:

Vendor

Labring

Status
Fastgpt
Vendor
CVE Published:
9 June 2025

What is CVE-2025-49131?

The FastGPT platform, which enables the deployment and execution of AI-driven workflows, contains a vulnerability in the sandbox environment (fastgpt-sandbox) prior to version 4.9.11. Due to insufficient isolation and overly permissive system calls, attackers could potentially escape the sandbox, allowing them to read and overwrite arbitrary files as well as bypass Python module import restrictions. The vulnerability has been addressed in version 4.9.11, where restrictions on allowed system calls were tightened, improving the overall security and isolation of user-generated code.

Affected Version(s)

FastGPT < 4.9.11

References

https://github.com/labring/FastGPT/security/advisories/GH...
x_refsource_CONFIRM
https://github.com/labring/FastGPT/pull/4958
x_refsource_MISC
https://github.com/labring/FastGPT/commit/bb810a43a1c7068...
x_refsource_MISC

CVSS V3.1

Score:
6.3
Severity:
MEDIUM
Confidentiality:
Low
Integrity:
Low
Availability:
Low
Attack Vector:
Network
Attack Complexity:
Low
Privileges Required:
Low
User Interaction:
None
Scope:
Unchanged

Timeline

  • Vulnerability published

  • Vulnerability Reserved

.