Fingerprinting Web Application Platforms by Variations in PNG Implementations
summary
Generates malformed PNG/MNG/jNG test images that exploit decoder differences and uses a Python tool to upload them and match response patterns to specific libraries (ImageMagick/libpng, GD, Python PIL, Java imaging, .NET/Mono, pngjs, ChunkyPNG, etc.). Demonstrates real-world matches against Amazon, Shopify, GitHub, Tumblr and Pinterest, produces segfaults and exceptions in some decoders, finds distinct MNG/jNG handling, and reports library IDs with a scoring algorithm — works whenever the site accepts PNG uploads.
AI-written summary · about
chapters (18)
- 0:00Introduction
- 0:35About me
- 0:54Agenda
- 1:42schizophrenic files
- 2:15example
- 3:59why
- 5:53pink
- 7:59Linked Files
- 8:49Libraries
- 9:46Images
- 10:01File Manager
- 10:26Java
- 10:53Web Application
- 11:30Amazon
- 12:10Photobucket
- 12:54Python
- 14:05Conclusion
- 16:19Results
related talks
22mSAD THUG...
15maCropalypse Now, in Search of the Missing PNG
35mProduct image optimization at Mister Spex
58mStegospolit - Exploit Delivery With Steganography and Polyglots
28mBuild your first Cyber Forensic Application
06mOvercoming the Challenges of Image Delivery
16mImage Processing in JavaScript
27mPlatPal: Detecting Malicious Documents with Platform Diversity