Home

Description

DocsGPT is a GPT-powered chat for documentation. From version 0.15.0 to before version 0.16.0, an attacker accessing both the official DocsGPT website or any local and public deployment, can craft a malicious payload bypassing the "MCP test" behavior to achieve arbitrary remote code execution (RCE). This issue has been patched in version 0.16.0.

PUBLISHED Reserved 2026-02-09 | Published 2026-04-29 | Updated 2026-05-06 | Assigner GitHub_M




CRITICAL: 10.0CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:H/VI:H/VA:H/SC:H/SI:H/SA:H

Problem types

CWE-77: Improper Neutralization of Special Elements used in a Command ('Command Injection')

Product status

>= 0.15.0, < 0.16.0
affected

References

github.com/...ocsGPT/security/advisories/GHSA-gcrq-f296-2j74 exploit

www.ox.security/...-vulnerabilities-across-the-ai-ecosystem/

www.ox.security/...mic-vulnerability-at-the-core-of-the-mcp/

github.com/...ocsGPT/security/advisories/GHSA-gcrq-f296-2j74

github.com/arc53/DocsGPT/releases/tag/0.16.0

cve.org (CVE-2026-26015)

nvd.nist.gov (CVE-2026-26015)

Download JSON