CVE-2026-5970

HIGH

FoundationAgents MetaGPT HumanEvalBenchmark/MBPPBenchmark check_solution code injection

Title source: cna

Description

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 (6)

[Vdb Entry, Technical Description] https://vuldb.com/vuln/356524

[Signature, Permissions Required] https://vuldb.com/vuln/356524/cti

[Third Party Advisory] https://vuldb.com/submit/791693

[Exploit] https://github.com/FoundationAgents/MetaGPT/issues/1942

[Patch] https://github.com/FoundationAgents/MetaGPT/pull/1988

[Product] https://github.com/FoundationAgents/MetaGPT/

Scores

CVSS v3 7.3

EPSS 0.0007

EPSS Percentile 22.1%

Attack Vector NETWORK

CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:L

CISA SSVC

Vulnrichment

Exploitation poc

Automatable yes

Technical Impact partial

Details

CWE

CWE-74 CWE-94

Status published

Products (3)

deepwisdom/metagpt < 0.8.1

FoundationAgents/MetaGPT 0.8.0

FoundationAgents/MetaGPT 0.8.1

Published Apr 09, 2026

Tracked Since Apr 09, 2026