One other flaw, CVE-2026-110186, permits unauthenticated distant attackers to execute arbitrary code by Mannequin Context Protocol (MCP) device enumeration. A 3rd vulnerability, CVE-2026-110193, which is rated as excessive severity, additionally permits for code and command execution on the server attributable to a problem in a third-party library utilized by the Agentic Assistant. Nonetheless, exploiting this flaw requires authentication with at the very least a low-privileged account.
The fourth vulnerability, CVE-2026-110192, is an data disclosure difficulty attributable to insufficiently protected API credentials. It’s rated as 6.5 on the CVSS scale.
The issues have an effect on Langflow OSS variations 1.0.0 to 1.12.8 and have been mounted within the newly launched 1.13.0 model, which is offered each from the PyPI repository and as a Docker picture.
