CVE-2026-33233Public exploit

Backend command execution via poisoned Redis pickle

Published May 19, 2026 · Updated May 19, 2026

Unsafe deserialization in Significant Gravitas AutoGPT 0.6.34 through 0.6.51 allows attackers with Redis write access to execute backend commands. The shared-cache read path in backend/util/cache.py passes unauthenticated Redis bytes directly to pickle.loads without a signature or schema check. Exploitation requires poisoning a shared-cache key and grants command execution in the backend container, exposing its data and service integrity.

CVSS severity7.6
High
EPSS probability0.23%
Next 30 days · Sep 16, 2026
Known exploitationUnconfirmed
Based on sourced intelligence
Hinoki checkNot available
Coverage for this vulnerability

See if you're affected

Explore vulnerability checks for your environment with Hinoki.

Book a demo

Summary

What happened

Unsafe deserialization in Significant Gravitas AutoGPT 0.6.34 through 0.6.51 allows attackers with Redis write access to execute backend commands. The shared-cache read path in backend/util/cache.py passes unauthenticated Redis bytes directly to pickle.loads without a signature or schema check. Exploitation requires poisoning a shared-cache key and grants command execution in the backend container, exposing its data and service integrity.

The record

CVE
CVE-2026-33233
Published
May 19, 2026
Updated
May 19, 2026
Vendor
Significant Gravitas
Product
AutoGPT
Classifications
CWE-502, CWE-345, CWE-94, T1059.004
Attack vector
adjacent
Privileges
Unavailable

Timeline

How it unfolded

  1. May 19, 2026CVE publishedPublication date reported by the CVE source.
  2. May 19, 2026Record updatedLatest update available in the CVE record.

Exploitability

Present is not the same as exploitable

Compare your product and version with the public record. A matching version still requires validation against your environment.

Is a vulnerable build present?

Compare these published version ranges with your installed build and any vendor patches.

  1. Affected versionversion=>= 0.6.34, < 0.6.52

What conditions does exploitation require?

Attack vectoradjacent
Required privilegesUnavailable in this record.

What is affected?

Significant Gravitas · AutoGPTversion=>= 0.6.34, < 0.6.52

Published CVSS scores

7.6GitHub, Inc.CVSS:3.1/AV:A/AC:H/PR:H/UI:N/S:C/C:H/I:H/A:H

CVSS describes severity. EPSS estimates exploitation probability.

Attacks

What attackers are doing with it

Daily unique IPs observed by Shadowserver honeypots for known exploited vulnerabilities (KEVs). Missing observations do not establish an absence of attacks.

Daily unique IPsNo honeypot observations are available for this CVE in the selected window.

No observations available

Sep 10, 2026Sep 16, 2026
Latest reporting daySep 16, 2026
Latest daily unique IPsUnavailable
Prior 30-day averageUnavailable
SourceShadowserver honeypots (KEV)
Vectoradjacent
PrivilegesUnavailable
Known exploitationUnconfirmed
Public exploitPublished

Weakness, pattern, technique

CWE-502Deserialization of Untrusted Data
CWE-345Insufficient Verification of Data Authenticity
CWE-94Improper Control of Generation of Code ('Code Injection')
T1059.004Unix Shell

Public exploit references

Labels summarize the accepted research assessment. They do not indicate a test against your environment.

Technologies

Your stack

See the directory against your own environment.

Your stack

Check the software in your environment

Book a demo to see how Hinoki identifies affected software and validates exploitability in your environment.

Book a demo