Prompt injection can lead to arbitrary Remote Code Execution (RCE) when user-supplied inputs, such as a URL, are not properly sanitized. In this example (./ffufai.py --max-extensions 5 -u https://hello.com`id`), the command injection occurs because the input (`id`) is being executed as part of the system command. This allows an attacker to run arbitrary commands on the system, like `id` (which displays user identity information), exploiting the input field's vulnerability to execute harmful actions. Proper input sanitization and validation can help prevent such attacks. Check the screenshot for a PoC.
