Deserialization Vulnerability in AutoGPT Workflow Automation Platform
CVE-2026-33233

7.6HIGH

Key Information:

Vendor

Significant-gravitas

Status
Autogpt
Vendor
CVE Published:
19 May 2026

What is CVE-2026-33233?

In versions 0.6.34 to 0.6.51 of AutoGPT, a workflow automation platform, unsafe deserialization occurs as the backend uses pickle.loads to process Redis cache bytes without performing integrity or authenticity checks. This weakness allows an attacker to manipulate shared-cache keys in Redis, potentially leading to arbitrary command execution within the backend container. This compromises the confidentiality, integrity, and availability of the system. This issue has been addressed in version 0.6.52.

Affected Version(s)

AutoGPT >= 0.6.34, < 0.6.52

References

https://github.com/Significant-Gravitas/AutoGPT/security/...
x_refsource_CONFIRM
https://github.com/Significant-Gravitas/AutoGPT/releases/...
x_refsource_MISC

CVSS V3.1

Score:
7.6
Severity:
HIGH
Confidentiality:
High
Integrity:
High
Availability:
High
Attack Vector:
Adjacent Network
Attack Complexity:
High
Privileges Required:
High
User Interaction:
None
Scope:
Changed

Timeline

  • Vulnerability published

  • Vulnerability Reserved

.