0xSebin

This is my running logbook for LLM02: Sensitive Information Disclosure - #2 in the OWASP Top 10 for LLM Applications (2026). Same format as the Prompt Injection log: a new hands-on writeup here eac...

This is my running logbook for LLM10: Improper Output Handling - #10 in the OWASP Top 10 for LLM Applications (2026), and the most “classic AppSec” of the lot. Same format as the Prompt Injection l...

PortSwigger’s Web Security Academy has a small, sharp topic called Web LLM attacks - four labs on attacking LLM-backed web apps: mapping what a model can reach, abusing its tools, and treating its ...

This is my running logbook for LLM03: Excessive Agency - #3 in the OWASP Top 10 for LLM Applications (2026), and the biggest climber on the 2026 list. Same format as the Prompt Injection log: a new...

This is my running logbook for LLM01: Prompt Injection - the top entry in the OWASP Top 10 for LLM Applications (2026). One vulnerability class, lots of different disguises. Rather than write a sep...

Every company I look at these days has quietly wired a language model into something that matters. A support bot that can read the ticketing system. A coding copilot with access to the repo. An “as...

“You don’t need a zero-day. You just need to look harder.” Every time I drop into an internal network during a pentest or a red team engagement, the story is always the same - the domain is a gol...