Which versions of Metagpt are affected by CVE-2026-5970?

FoundationAgents MetaGPT versions up to 0.8.1 contain an unauthenticated remote code execution vulnerability in benchmark evaluation components that allows attackers to execute arbitrary code without…. A patch is available.

Is there a public PoC exploit available for CVE-2026-5970?

Yes, a public proof-of-concept exploit is available for CVE-2026-5970. Prioritise patching immediately. EPSS: 0.1%.

How to fix or mitigate CVE-2026-5970?

Within 24 hours: Identify all systems running FoundationAgents MetaGPT versions 0.8.1 and earlier; isolate affected instances from production networks if upgrade cannot be completed immediately. Within 7 days: Upgrade all instances to the latest patched version released by FoundationAgents (confirm version >0.8.1 with vendor advisory). Within 30 days: Conduct forensic review of MetaGPT system logs and audit data access for unauthorized activity during the vulnerability window; implement network segmentation to restrict MetaGPT component access.

Metagpt CVE-2026-5970

Q: What is the CVSS score of CVE-2026-5970?

CVE-2026-5970 has a CVSS 4.0 base score of 5.5 (Medium). CVSS vector: CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:L/VI:L/VA:L/SC:N/SI:N/SA:N/E:P/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X. EPSS exploitation probability: 0.1%.

EUVD-2026-20984 MEDIUM

Code Injection (CWE-94)

2026-04-09 VulDB

GHSA-g977-h85w-h2xj

RCE Code Injection Metagpt

5.5

CVSS 4.0 · NVD

Severity by source

NVD PRIMARY

5.5 MEDIUM

CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:L/VI:L/VA:L/SC:N/SI:N/SA:N/E:P/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X

Primary rating from NVD · only source for this CVE.

CVSS VectorNVD

CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:L/VI:L/VA:L/SC:N/SI:N/SA:N/E:P/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X

Attack Vector

Network

Attack Complexity

Low

Privileges Required

None

User Interaction

None

Scope

Lifecycle Timeline

CVSS changed

Apr 29, 2026 - 01:11 NVD

6.9 (MEDIUM) 5.5 (MEDIUM)

PoC Detected

Apr 09, 2026 - 18:17 vuln.today

Public exploit code

EUVD ID Assigned

Apr 09, 2026 - 17:45 euvd

EUVD-2026-20984

Analysis Generated

Apr 09, 2026 - 17:45 vuln.today

Patch released

Apr 09, 2026 - 17:45 nvd

Patch available

CVE Published

Apr 09, 2026 - 17:00 nvd

MEDIUM 6.9

DescriptionCVE.org

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.

AnalysisAI

Code injection in FoundationAgents MetaGPT versions up to 0.8.1 allows unauthenticated remote attackers to execute arbitrary code through the check_solution function in HumanEvalBenchmark and MBPPBenchmark components. The vulnerability requires no user interaction and enables unauthorized access, data modification, and service disruption. Publicly available exploit code exists, disclosed via GitHub issue #1942. Vendor has not responded to early disclosure via pull request #1988 at time of analysis.

Technical ContextAI

CWE-94 code injection flaw in check_solution function within benchmark evaluation modules. Unsanitized input to code execution context permits arbitrary code injection. CVSS vector AV:N/AC:L/PR:N confirms network-accessible attack with low complexity and no authentication requirement. Affects FoundationAgents MetaGPT cpe:2.3:a:foundationagents:metagpt:*:*:*:*:*:*:*:* through version 0.8.1.

RemediationAI

Upstream fix available (PR/commit); released patched version not independently confirmed. Vendor notified via GitHub pull request #1988 (https://github.com/FoundationAgents/MetaGPT/pull/1988) but has not issued formal response or release at time of analysis. Organizations should monitor vendor repository for patched release incorporating PR #1988 fix. Until patched version available, implement strict input validation on all user-supplied data passed to HumanEvalBenchmark and MBPPBenchmark check_solution functions. Consider disabling affected benchmark components if not operationally required. Apply network segmentation to limit remote access to MetaGPT instances. Review VulDB advisory https://vuldb.com/vuln/356524 and GitHub issue #1942 for additional technical details and potential workarounds.