The vulnerability is an improper input validation flaw in IBM Langflow OSS 1.0.0 through 1.10.1’s PythonREPL sandbox implementation. Based on the description, it is inferred that the flaw allows an attacker to bypass the sandbox and execute arbitrary code, which can be used to read or exfiltrate sensitive data stored on the system. The weakness is classified as CWE‑94, indicating a code injection issue that compromises confidentiality.

IBM Langflow OSS, versions 1.0.0 through 1.10.1 are affected.

The CVSS score of 9.9 denotes a critical level of risk. The EPSS score of < 1% indicates a very low but nonzero exploitation probability, yet the lack of mitigation and the high CVSS suggest that if an attack vector is usable, exploitation is likely. The vulnerability is not listed in CISA’s KEV catalog, yet the severity warrants caution. Based on the description, it is inferred that the likely attack vector is the input to the PythonREPL feature, most plausibly supplied via the application’s API or web interface; the attacker would need to provide malicious code that bypasses sandbox controls, leading to data exposure.