CVE-2026-5970CRITICAL 9.8EPSS p30.4%

CVE-2026-5970CVE-2026-5970

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.

Scoring

CVSS 3.19.8 (CRITICAL)
VectorCVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H
EPSS0.39% probability of exploitation · percentile 30.4% · 2026-06-19T12:03:05Z
Published2026-04-09
Last modified2026-04-29

Underlying weaknesses· 2

CWE-74CWE-94

References

  1. https://github.com/FoundationAgents/MetaGPT/
  2. https://github.com/FoundationAgents/MetaGPT/issues/1942
  3. https://github.com/FoundationAgents/MetaGPT/pull/1988
  4. https://vuldb.com/submit/791693
  5. https://vuldb.com/vuln/356524
  6. https://vuldb.com/vuln/356524/cti

2

TypeTargetConfidenceTier
WeaknessImproper Neutralization of Special Elements in Output Used by a Downstream Component ('Injection')cwe-740%live
WeaknessImproper Control of Generation of Code ('Code Injection')cwe-940%live

Related by meaning· 6

Nearest entities by semantic similarity across the cs-graph corpus.

CVE
CVE-2026-5971
CVE
CVE-2026-5974
CVE
CVE-2026-6110
CVE
CVE-2026-5972
CVE
CVE-2026-5973
CVE
CVE-2026-10566
Sourced from NVD + FIRST.org EPSS. Curated for EU compliance use cases by Adam Lundqvist, Founder at SQUR.