CVE-2025-14009

The core issue lies within the unzipiter function in nltk/downloader.py, which utilizes zipfile.extractall() without proper security checks. An attacker can craft a specially designed zip archive containing malicious Python files, such as init.py. When a user downloads and extracts this archive using NLTK, these malicious files will be automatically executed upon import, granting the attacker arbitrary code execution privileges. This effectively bypasses standard security measures, as NLTK inherently trusts downloaded packages. The blast radius is significant, potentially impacting any system running vulnerable versions of NLTK and processing downloaded data.

Researchers and developers utilizing NLTK for natural language processing tasks are at significant risk. Specifically, those using older versions of NLTK (≤3.9b1) or those who automatically process downloaded data without proper validation are particularly vulnerable. Shared hosting environments where multiple users may be utilizing NLTK are also at increased risk.

• python / supply-chain:

import os
import zipfile

def check_nltk_download_path(download_dir):
    for filename in os.listdir(download_dir):
        if filename.endswith('.zip'):
            try:
                with zipfile.ZipFile(os.path.join(download_dir, filename), 'r') as zip_ref:
                    for member in zip_ref.infolist():
                        if not member.filename.startswith('nltk/'):
                            print(f"Suspicious file found in zip: {member.filename}")
            except Exception as e:
                print(f"Error processing zip file: {e}")

# Example usage (replace with actual download directory)
nltk_download_dir = '/path/to/nltk/downloads'
check_nltk_download_path(nltk_download_dir)

• generic web: Check NLTK download directories for unexpected files or scripts. • generic web: Monitor system logs for unusual Python process execution after NLTK downloads.

1. 2026-02-18Disclosure

   disclosure

2. 2026-02-18Patch

   patch

1. 予約済み2025-12-04
2. 公開日2026-02-18
3. 更新日2026-02-25
4. EPSS 更新日2026-03-23

The primary mitigation is to immediately upgrade to NLTK version 3.9.3 or later. If upgrading is not immediately feasible due to compatibility issues or breaking changes, consider temporarily disabling the NLTK downloader and manually verifying the integrity of any downloaded data. Implement strict file integrity checks on all downloaded packages before processing them. While a WAF or proxy cannot directly address this vulnerability, they can be configured to block downloads from untrusted sources. There are no specific Sigma or YARA rules readily available for this vulnerability, but monitoring for unusual Python process execution after NLTK downloads is recommended.