CVE Intelligence
Analysis AI-assisted

CVE-2026-0761: Foundation Agents MetaGPT Code Injection in actionoutput_str_to_mapping Leading to Remote Code Execution

CVE-2026-0761 is a code injection flaw in the actionoutput_str_to_mapping function of Foundation Agents MetaGPT, classified as CWE-94 in the record. According to the Zero Day Initiative advisory, a remote attacker without authentication may execute arbitrary code in the context of the service account. The record lists a CVSS base score of 9.8 and no confirmed exploitation in the sources consulted.

Published Oct 24, 2026 · Updated Oct 24, 2026 · Ler em português

Overview

The Zero Day Initiative advisory describes a code injection flaw in the actionoutput_str_to_mapping function of Foundation Agents MetaGPT. The record maps the weakness to CWE-94 (Code Injection). The function reportedly uses a user-supplied string to execute Python code without proper validation.

Impact

  • The advisory states that remote attackers can execute arbitrary code without authentication.
  • Code would run in the context of the service account running MetaGPT.
  • The record gives a CVSS base score of 9.8 (critical) and an SSVC technical impact of total.

Affected products and versions

  • The record lists Deepwisdom Metagpt and FoundationAgents MetaGPT, both at version 0.8.1.
  • Whether other versions are affected is unknown from the sources consulted.

Exploitation status

  • CISA KEV does not list this CVE.
  • No confirmed exploitation in the wild appears in the sources consulted.
  • The SSVC exploitation field is recorded as none. The EPSS probability is 0.02413 with a percentile of 0.84005.

Remediation and mitigation

  • The record contains no fixed version. This absence does not confirm that no fix exists.
  • Check the ZDI advisory and vendor channels for updates and patched releases.
  • Until a fix is confirmed, limit network exposure of MetaGPT services, run them with least privilege, and review which inputs can reach the affected function.

Detection guidance

  • Review logs for unexpected child processes spawned by MetaGPT service processes.
  • Monitor agent hosts for unusual outbound connections or changes to files they should not write.

Limitations of the available information

  • The record includes no vendor statement, patch, or fixed version.
  • Preconditions, deployment configurations, and the exact version boundaries are not specified.

Frequently asked questions

What is the core weakness in CVE-2026-0761?

The record classifies it as CWE-94 (Code Injection). The advisory says the actionoutput_str_to_mapping function uses a user-supplied string to execute Python code without adequate validation.

Is authentication required to exploit this issue?

According to the advisory text in the record, no authentication is required.

Is this CVE known to be exploited in the wild?

CISA KEV does not list it, and the sources consulted show no confirmed exploitation in the wild.

Is a fixed version available?

The record lists no fixed version. That does not prove that no fix exists, so check the ZDI advisory and vendor channels for updates.

This analysis was drafted with AI assistance from verified source data and automatically checked against it.