Authorization Bypass Vulnerability in AgentGPT by Geo Chen
CVE-2026-65699

2.3LOW

Key Information:

Vendor

Reworkd

Status
Agentgpt
Vendor
CVE Published:
23 July 2026

Badges

👾 Exploit Exists🟡 Public PoC

What is CVE-2026-65699?

AgentGPT version 1.0.0 has a critical flaw that permits authenticated users to bypass authorization checks by manipulating request parameters. This vulnerability allows users to attach tasks to agent runs belonging to other users without verifying ownership, posing a significant risk of task history corruption and excessive resource consumption against innocent users' runs. Attackers who exploit this weakness can gain unauthorized access to sensitive operations, leading to potential financial impacts and undermined user trust.

Affected Version(s)

AgentGPT 0 <= 1.0.0

Exploit Proof of Concept (PoC)

PoC code is written by security researchers to demonstrate the vulnerability can be exploited. PoC code is also a key component for weaponization which could lead to ransomware.

CVE-2026-65699 - Proof of Concept

References

https://github.com/geo-chen/oss/blob/main/AgentGPT.md
technical-descriptionexploit
https://www.vulncheck.com/advisories/agentgpt-authorizati...
third-party-advisory

CVSS V4

Score:
2.3
Severity:
LOW
Confidentiality:
None
Integrity:
Low
Availability:
Low
Attack Vector:
Network
Attack Complexity:
High
Attack Required:
None
Privileges Required:
Undefined
User Interaction:
None

Timeline

  • 🟡

    Public PoC available

  • 👾

    Exploit known to exist

  • Vulnerability published

  • Vulnerability Reserved

Credit

George Chen
.