# CVE-2026-10566: FoundationAgents MetaGPT unsafe deserialization in Message.check_instruct_content

Source: https://junglewise.ai/threats/cve-2026-10566-foundationagents-metagpt-unsafe-deserialization-in-message-check  
Publisher: Junglewise Threat Intelligence, https://junglewise.ai/threats  
Published: 2026-06-02. Updated: 2026-07-09  
License: CC BY 4.0 with a link back

CVE-2026-10566 · Severity: medium · CVSS 5.3 · Published 2026-06-02

FoundationAgents MetaGPT is a multi-agent framework used to build AI-driven software applications. A security flaw in how the system handles message data allows an attacker with local access to execute unauthorized commands on the host machine. This could lead to a full system compromise or unauthorized access to sensitive data processed by the AI agents.

Technologies: [metagpt (PyPI)](https://junglewise.ai/threats/technologies/pypi-metagpt). Vendors: [PyPI](https://junglewise.ai/threats/vendors/pypi).

## Quick answers

**Is CVE-2026-10566 exploited in the wild?** No. Junglewise Threat Intelligence has no report of CVE-2026-10566 being exploited in the wild.

**How severe is CVE-2026-10566?** Medium severity, with a CVSS score of 5.3, published 2 June 2026.

**Which products and versions are affected?** metagpt (PyPI) ECOSYSTEM:introduced=0,last_affected=0.8.2.

**Is there a fix for CVE-2026-10566?** No fixed version is recorded yet. Check the vendor's advisory in the references below. If you run it, Junglewise drafts a mitigation that fits your setup and applies it once you approve, with senior security engineers behind it.

**How is CVE-2026-10566 attacked?** It is attacked with local access, in a software library.

**Am I affected by CVE-2026-10566?** Check whether you run metagpt (PyPI) at an affected version. Junglewise does that check for you around the clock: connect your code, cloud and domains, and it matches new vulnerabilities like this one against what you actually run and tells you when one reaches you.

## Technical details

A deserialization vulnerability exists in FoundationAgents MetaGPT up to version 0.8.2 within the `Message.check_instruct_content` function in `metagpt/schema.py`. The root cause is the unsafe use of the `eval()` function in `metagpt/utils/serialize.py:actionoutput_str_to_mapping()`, which processes attacker-controlled strings during the mapping of message arguments. An attacker with the ability to influence serialized Message data (e.g., through local storage or recovery paths) can achieve arbitrary Python code execution. While the attack is restricted to local execution, a public exploit has been released. As of the advisory date, no official patch has been released by the maintainers.

## Affected products

- FoundationAgents MetaGPT <= 0.8.2

## Identifiers

- CVE-2026-10566
- GHSA-3c3g-7hwg-9qmr
- PYSEC-2026-2637

## Timeline

- 2026-05-13: disclosed: Issue reported to the maintainers via GitHub issue #2038
- 2026-06-02: advisory: Published to the GitHub Advisory Database and NVD

## References

- [https://github.com/FoundationAgents/MetaGPT/issues/2038](https://github.com/FoundationAgents/MetaGPT/issues/2038)
- [https://github.com/FoundationAgents/MetaGPT](https://github.com/FoundationAgents/MetaGPT)
- [https://vuldb.com/cve/CVE-2026-10566](https://vuldb.com/cve/CVE-2026-10566)
- [https://vuldb.com/submit/828301](https://vuldb.com/submit/828301)
- [https://vuldb.com/vuln/367673](https://vuldb.com/vuln/367673)

## Related threats

- [FoundationAgents MetaGPT SSRF in decode_image](https://junglewise.ai/threats/cve-2026-6111-foundationagents-metagpt-ssrf-in-decode-image)
- [FoundationAgents MetaGPT code injection in Tree-of-Thought Solver](https://junglewise.ai/threats/cve-2026-6110-foundationagents-metagpt-code-injection-in-tree-of-thought-solver)
- [FoundationAgents MetaGPT RCE via CSRF in Mineflayer HTTP API](https://junglewise.ai/threats/cve-2026-6109-foundationagents-metagpt-rce-via-csrf-in-mineflayer-http-api)
- [FoundationAgents MetaGPT OS command injection in Bash tool](https://junglewise.ai/threats/cve-2026-5974-foundationagents-metagpt-os-command-injection-in-bash-tool)
- [FoundationAgents MetaGPT OS command injection in get_mime_type](https://junglewise.ai/threats/cve-2026-5973-foundationagents-metagpt-os-command-injection-in-get-mime-type)