jwt_tool.py is a toolkit for validating, forging and cracking JWTs (JSON Web Tokens).
Its functionality includes:
- Checking the validity of a token
- Testing for the RS/HS256 public key mismatch vulnerability
- Testing for the alg=None signature-bypass vulnerability
- Testing the validity of a secret/key/key file
- Identifying weak keys via a High-speed Dictionary Attack
- Forging new token header and payload values and creating a new signature with the key or via another attack method
This tool is written for pentesters, who need to check the strength of the tokens in use, and their susceptibility to known attacks.
It may also be useful for developers who are using JWTs in projects, but would like to test for stability and for known vulnerabilities, when using forged tokens.
This tool is written natively in Python 2.x using the common libraries.
Customised wordlists are recommended for the Dictionary Attack option.
As a speed reference, an Intel i5 laptop can test ~1,000,000 passwords per second on HMAC-SHA256 signing. YMMV.
Installation is just a case of downloading the
jwt_tool.py file (or
git cloneing the repo).
chmod the file too if you want to add it to your $PATH and call it from anywhere.)
$ python jwt_tool.py <JWT> (filename)
The first argument should be the JWT itself, followed by a filename/filepath (for cracking the token, or for use as a key file).
$ python jwt_tool.py
The toolkit will validate the token and list the header and payload values.
It will then provide a menu of your available options.
Note: signing the token is currently supported using HS256, HS384, HS512 algorithms
Input is in either standard or url-safe JWT format, and the resulting tokens are output in both formats for your ease of use.
- A great intro to JWTs – https://jwt.io/introduction/
- A lot of the inspiration for this tool comes from the vulnerabilities discovered by Tim McLean.
Check out his blog on JWT weaknesses here: https://auth0.com/blog/critical-vulnerabilities-in-json-web-token-libraries/
- My introduction to using this toolkit, and a bit of the history behind it can be found on my blog – https://www.ticarpi.com/introducing-jwt-tool/
- A whole bunch of exercises (7 at time of writing) for testing JWT vulnerabilities are provided by Pentesterlab. I’d highly recommend a PRO subscription if you are interested in Web App Pentesting.
- JWT (alg=None vulnerability) exercise
- JWT_II (RS/HS256 public key mismatch vulnerability) exercise
- JWT_III (key-id header field non-sanitisation vulnerability) exercise
- and just head on over to https://pentesterlab.com/exercises to search for the others!
Regex for finding JWTs in Burp Search
(make sure ‘Case sensitive’ and ‘Regex’ options are ticked)
[= ]ey[A-Za-z0-9_-]*\.[A-Za-z0-9._-]* – url-safe JWT version
[= ]ey[A-Za-z0-9_\/+-]*\.[A-Za-z0-9._\/+-]* – all JWT versions (higher possibility of false positives)