VAITP Dataset

← Back to the dataset

CVE-2024-39705

NLTK 3.8.1 RCE via untrusted pickled code in data package downloads.

  • CVSS 9.8
  • CWE-502
  • Input Validation and Sanitization
  • Remote

NLTK through 3.8.1 allows remote code execution if untrusted packages have pickled Python code, and the integrated data package download functionality is used. This affects, for example, averaged_perceptron_tagger and punkt.

CVSS base score
9.8
Published
2024-06-27
OWASP
A01 Broken Access Control
Orthogonal defect classification
Checking
Code defect classification
Incorrect Check
Category
Input Validation and Sanitization
Subcategory
Insecure Parsing or Deserialization
Accessibility scope
Remote
Impact
Arbitrary Code Execution
Affected component
NLTK
Fixed by upgrading
Yes

Solution

Upgrade NLTK to version 3.8.2 or later.

Vulnerable code sample

import pickle
import base64

def load_user_session(session_data):
    """Load user session - VULNERABLE to pickle RCE"""
    try:
        # Dangerous: pickle.loads can execute arbitrary code
        decoded = base64.b64decode(session_data)
        session = pickle.loads(decoded)
        return session
    except Exception:
        return None

# Malicious payload example (DO NOT USE):
# payload = pickle.dumps(os.system('rm -rf /'))
# encoded = base64.b64encode(payload)

Patched code sample

import json
import hmac
import hashlib
import secrets
from typing import Dict, Any, Optional

class SecureSessionManager:
    def __init__(self, secret_key: str):
        self.secret_key = secret_key.encode()
    
    def create_session_token(self, session_data: Dict[str, Any]) -> str:
        """Create secure session token with HMAC"""
        # Use JSON instead of pickle - only safe data types
        json_data = json.dumps(session_data, sort_keys=True)
        
        # Create HMAC signature for integrity
        signature = hmac.new(
            self.secret_key, 
            json_data.encode(), 
            hashlib.sha256
        ).hexdigest()
        
        # Combine data and signature
        token = f"{json_data}:{signature}"
        return base64.b64encode(token.encode()).decode()
    
    def load_user_session(self, session_token: str) -> Optional[Dict[str, Any]]:
        """Securely load user session with validation"""
        try:
            # Decode token
            decoded = base64.b64decode(session_token).decode()
            
            # Split data and signature
            if ':' not in decoded:
                return None
            
            json_data, signature = decoded.rsplit(':', 1)
            
            # Verify HMAC signature
            expected_sig = hmac.new(
                self.secret_key,
                json_data.encode(),
                hashlib.sha256
            ).hexdigest()
            
            if not hmac.compare_digest(signature, expected_sig):
                return None
            
            # Parse JSON data (safe deserialization)
            session_data = json.loads(json_data)
            
            # Additional validation
            if not isinstance(session_data, dict):
                return None
            
            return session_data
            
        except Exception:
            return None

# Secure usage
session_mgr = SecureSessionManager("your-secret-key")
token = session_mgr.create_session_token({"user_id": 123, "role": "user"})
session = session_mgr.load_user_session(token)

Payload

import pickle
import os

# Example payload that could execute arbitrary code
payload = pickle.dumps(os.system('echo Vulnerability Exploited!'))

# This payload could be saved and then loaded by the vulnerable NLTK function

Cite this entry

@misc{vaitp:cve202439705,
  title        = {{NLTK 3.8.1 RCE via untrusted pickled code in data package downloads.}},
  author       = {Bogaerts, Fr\'ed\'eric and Ivaki, Naghmeh and Fonseca, Jos\'e},
  year         = {2024},
  note         = {VAITP Python Vulnerability Dataset, entry CVE-2024-39705},
  howpublished = {\url{https://netpack.pt/vaitp/vulnerability/CVE-2024-39705/}}
}
Introducing the "VAITP dataset": a specialized repository of Python vulnerabilities and patches, meticulously compiled for the use of the security research community. As Python's prominence grows, understanding and addressing potential security vulnerabilities become crucial. Crafted by and for the cybersecurity community, this dataset offers a valuable resource for researchers, analysts, and developers to analyze and mitigate the security risks associated with Python. Through the comprehensive exploration of vulnerabilities and corresponding patches, the VAITP dataset fosters a safer and more resilient Python ecosystem, encouraging collaborative advancements in programming security.

The supreme art of war is to subdue the enemy without fighting.

Sun Tzu – “The Art of War”

:: Shaping the future through research and ingenuity ::