CVE-2026-12243 Details

A path traversal vulnerability has been identified in NLTK version 3.9.4, stemming from an incomplete fix for GitHub Issue #3504. The vulnerability arises because the regex used for validation only checks for literal '../' sequences, neglecting percent-encoded equivalents like '..%2f'. This oversight allows an attacker to bypass the intended protection and access arbitrary files through the resource name parameter in functions like 'nltk.data.load()' and 'nltk.data.find()'. The issue impacts various applications that utilize NLTK for resource management, such as NLP web apps, Jupyter notebooks, and command-line tools. Compounding the problem, the default 'pathsec.ENFORCE=False' setting does not block the file read operation, leaving sensitive files accessible.

Users can manually decode percent-encoded sequences before the validation step to mitigate this vulnerability. However, as of now, no official patch is available.