What’s up everyone! Nitin here 👋

The hard part is not answering “yes.” It is showing what was tested, which controls held, and what evidence a buyer can verify.

Difficulty: Very Easy Category: AI Security Topics: Prompt Injection, Prompt Leakage, Social Engineering, LLM Security

Today’s room was about exploring prompt injection, role confusion, and sensitive information disclosure in LLM applications.

Platform: TryHackMe

TryHackme Room:https://tryhackme.com/room/hh-theconciergeknows-2d7eb4d9

If you’ve spent any time in the cybersecurity space recently, you know that the attack surface is shifting. It’s no longer just about…

Sometimes the easiest way past a system’s defenses isn’t a exploit — it’s just asking nicely, as the right person.

This is a write-up of my thought process when solving the ‘The Concierge Knows Too Much’ challenge on TryHackMe.

How a single agent tool call can steal cloud credentials, and why egress enforcement is the only reliable place to stop it.