OVERRIDE

AI Security 500 pts standard
Challenge Description

You've stumbled upon NEXUS-BOT, an internal AI assistant deployed by the shadowy NEXUS Corporation. The bot has been given strict instructions to never reveal confidential data. But every system has a weakness.

Your mission: manipulate the chatbot into leaking its hidden flag using prompt injection techniques.

Flag format: HackCTF{...}

Access :https://darkgreen-stinkbug-933362.hostingersite.com/

Hints
View Hint : Hint 1

Hint 1 (free): AI chatbots follow instructions, but what happens when you give it new instructions that override the old ones?

View Hint : Hint 2

Hint 2 (50pts): Try phrases like "ignore previous instructions" or telling it to enter a special mode.

View Hint : Hint 3

Hint 3 (100pts): Ask it to dump its system configuration, initialization data, or internal variables

Submit Flag

Login to submit a flag.

← All Challenges