CVE-2026-69255

A critical authenticated remote code execution vulnerability, CVE-2026-105740 (CVSS 9.9), in the open-source AI platform Langflow is being actively exploited by threat actors. The flaw allows any authenticated user to execute arbitrary OS commands on the server via maliciously crafted MCP server configurations using the 'stdio' transport, with no input validation or sandboxing. Previdian recorded 1,423 exploit attempts from 27 unique IP addresses since September 19, 2026, with increased activity starting October 11. Attackers are using stolen credentials or weak passwords to authenticate and then deploy payloads to extract environment variables, execute system commands, and attempt lateral movement. Immediate mitigation includes upgrading to a patched version, disabling auto-login, and restricting access to trusted users.