CVE-2026-5970

A vulnerability was detected in FoundationAgents MetaGPT up to 0.8.1. This affects the function check_solution of the component HumanEvalBenchmark/MBPPBenchmark. Performing a manipulation results in code injection. The attack may be initiated remotely. The exploit is now public and may be used. The project was informed of the problem early through a pull request but has not reacted yet.
References
Link Resource
https://github.com/FoundationAgents/MetaGPT/ Product
https://github.com/FoundationAgents/MetaGPT/issues/1942 Exploit Issue Tracking
https://github.com/FoundationAgents/MetaGPT/pull/1988 Issue Tracking Patch
https://vuldb.com/submit/791693 Exploit Third Party Advisory VDB Entry
https://vuldb.com/vuln/356524 Third Party Advisory VDB Entry
https://vuldb.com/vuln/356524/cti Permissions Required VDB Entry
Configurations

Configuration 1 (hide)

cpe:2.3:a:deepwisdom:metagpt:*:*:*:*:*:*:*:*

History

No history.

Information

Published : 2026-04-09 18:17

Updated : 2026-06-17 10:59


NVD link : CVE-2026-5970

Mitre link : CVE-2026-5970

CVE.ORG link : CVE-2026-5970


JSON object : View

Products Affected

deepwisdom

  • metagpt
CWE
CWE-74

Improper Neutralization of Special Elements in Output Used by a Downstream Component ('Injection')

CWE-94

Improper Control of Generation of Code ('Code Injection')