Vulnerability Details 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.
Exploit prediction scoring system (EPSS) score
EPSS Score 0.001
EPSS Ranking 22.1%
CVSS Severity
CVSS v3 Score 7.3
CVSS v2 Score 7.5
References
Products affected by CVE-2026-5970
-
cpe:2.3:a:deepwisdom:metagpt:0.1.0
-
cpe:2.3:a:deepwisdom:metagpt:0.2.0
-
cpe:2.3:a:deepwisdom:metagpt:0.2.1
-
cpe:2.3:a:deepwisdom:metagpt:0.3.0
-
cpe:2.3:a:deepwisdom:metagpt:0.4.0
-
cpe:2.3:a:deepwisdom:metagpt:0.5.0
-
cpe:2.3:a:deepwisdom:metagpt:0.5.1
-
cpe:2.3:a:deepwisdom:metagpt:0.5.2
-
cpe:2.3:a:deepwisdom:metagpt:0.6.0
-
cpe:2.3:a:deepwisdom:metagpt:0.6.2
-
cpe:2.3:a:deepwisdom:metagpt:0.6.3
-
cpe:2.3:a:deepwisdom:metagpt:0.6.4
-
cpe:2.3:a:deepwisdom:metagpt:0.6.6
-
cpe:2.3:a:deepwisdom:metagpt:0.7.0
-
cpe:2.3:a:deepwisdom:metagpt:0.7.1
-
cpe:2.3:a:deepwisdom:metagpt:0.7.2
-
cpe:2.3:a:deepwisdom:metagpt:0.7.3
-
cpe:2.3:a:deepwisdom:metagpt:0.7.4
-
cpe:2.3:a:deepwisdom:metagpt:0.7.6
-
cpe:2.3:a:deepwisdom:metagpt:0.7.7
-
cpe:2.3:a:deepwisdom:metagpt:0.8.0
-
cpe:2.3:a:deepwisdom:metagpt:0.8.1