Langflow flaw exploited to steal OpenAI, AWS keys

Threat actors are exploiting CVE-2026-0768, an unauthenticated remote code execution vulnerability in Langflow versions 1.4.2 and earlier, to harvest credentials, tokens, and keys. VulnCheck observed 360 attacks, primarily from Russia, targeting U.K. honeypots. The flaw resides in the code validator of the custom component editor.