Server-Side Browsing Considered Harmful
summary
Shows concrete SSRF attack chains: using dotless decimal and octal IP encodings, DNS rebinding and a dynamic HTTP redirect endpoint to bypass blacklists; targets include loopback, cloud metadata multicast and internal nets. Uses DNSChef, Burp Intruder, webbins and a custom nodejs port scanner and proxy; the scanner and Redis proxy work and yield real compromises. Reports list payouts ($500, $750, $3K, $5K, $15K, $20K) and specific library bypasses like IPy that remain exploitable.
AI-written summary · about
chapters (11)
- 0:00Introduction and scope
- 2:52Identifying attack vectors
- 8:11Target destinations
- 12:09Loopback and metadata
- 20:02Internal network attacks
- 24:38Bypassing blacklists
- 29:31IP and DNS evasion
- 37:12Case studies and bounties
- 47:18Advanced exploitation
- 57:10Facebook and conclusion
- 1:00:00Toolbox and QA
related talks
42mModern web application bugs
58mModern web application bugs
33mOffensive Javascript Techniques for Red Teamers
56mModern Web Vulnerabilities 2020
35mServer-Side Browsers: Exploring the Web’s Hidden Attack Surface
45mHTTP Desync Attacks: Smashing Into The Cell Next Door
28mOffensive and Defensive Strategies for Client-Side JS Security
47mLive Hacking: Breaking into Your Web App
