Executive brief
IBM Langflow Desktop, a tool used for building and testing AI workflows, contains a vulnerability that allows an attacker to run unauthorized commands on the host system. By exploiting this flaw, an authenticated user could steal sensitive information like API keys and database credentials or gain full control over the application environment. Organizations should update to version 1.9.0 immediately to prevent potential data breaches or internal network attacks.
Technical details
A code injection vulnerability (CWE-94) exists in the IBM Langflow Desktop code validation functionality. The '/api/v1/validate/code' endpoint utilizes Python's exec() function to process user-supplied input but fails to properly account for decorator execution during function definition parsing. An authenticated attacker can inject malicious decorators that execute immediately during the validation process. This allows for arbitrary command execution on the server with the privileges of the Langflow backend service. The vulnerability is resolved in version 1.9.0.
Affected products
- IBM Langflow Desktop 1.0.0 - 1.8.4
Timeline
- 2026-04-28: advisory: Initial publication by IBM
- 2026-04-30: disclosed: CVE-2026-6543 published
- 2026-05-11: other: NVD initial analysis completed