High severity7.3NVD Advisory· Published Apr 9, 2026· Updated Apr 29, 2026
CVE-2026-5970
CVE-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.
Affected packages
Versions sourced from the GitHub Security Advisory.
| Package | Affected versions | Patched versions |
|---|---|---|
metagptPyPI | <= 0.8.1 | — |
Affected products
1Patches
0No patches discovered yet.
Vulnerability mechanics
AI mechanics synthesis has not run for this CVE yet.
References
7- github.com/FoundationAgents/MetaGPT/pull/1988nvdIssue TrackingPatchWEB
- github.com/FoundationAgents/MetaGPT/issues/1942nvdIssue TrackingExploitWEB
- vuldb.com/submit/791693nvdExploitThird Party AdvisoryVDB EntryWEB
- github.com/advisories/GHSA-g977-h85w-h2xjghsaADVISORY
- nvd.nist.gov/vuln/detail/CVE-2026-5970ghsaADVISORY
- vuldb.com/vuln/356524nvdThird Party AdvisoryVDB EntryWEB
- vuldb.com/vuln/356524/ctinvdPermissions RequiredVDB EntryWEB
News mentions
0No linked articles in our index yet.