5PwNDCv.png ap64jNt.png DhWShYz.png ZgrT7eX.png
PillowFight uses advanced AI/MLRegressionLearning to combine two images of your choosing
*note to investors this is not techically true at the moment we're using a python library but please give us money and we'll deliver it we promise.
Press the Start
button on the top-right to begin this challenge.
Upon accessing the website, we find a form that allows uploading two images and combining them. There’s also a hint about API documentation available at /swagger
. Exploring this documentation reveals an endpoint, /combine
, where we can specify images along with an eval_command
parameter for custom operations. Let’s test it by attempting to combine two images:
Here's what we've got:
Let’s try using a different custom eval_command
instead of convert()
in the previous formData, maybe can we do some command injection.
This attempt results in a "Bad Request", with the following response:
It appears that the server expects an image output with a .save()
attribute. So, what if we raise an exception instead to see the error response? Let's try this:
Bingo! The flag appears in the response body as part of the error message.
The flag is: flag{b6b62e6c5cdfda3b3a8b87d90fd48d01}