NEWPractise supported commands in browser simulations, alongside the lesson.Try a browser practice exercise.Try a lab♥ Codelivly is 100% free, no paywall. If it helped you, chip in to keep it running.100% free. Help keep it so.Support us
Linux Fundamentals for Cybersecurity
Linux is the backbone of modern cybersecurity. Whether you are defending enterprise servers, analyzing malware, or launc
ViewRecon & Enumeration Tools
Reconnaissance is the foundation of every successful penetration test. Amateurs treat it as a box-ticking exercise, runn
ViewCredential Attacks
Amateurs think credential attacks are just about guessing "Password123" or running a generic brute-force script until th
ViewNetwork Services Administration
This comprehensive learning path bridges the gap between foundational networking and enterprise infrastructure security.
ViewComputer Fundamentals
Before you can hack, defend, or analyze malware, you must understand the machines you are working with. The Computer Fun
ViewNetworking Fundamentals
You cannot secure a network if you do not understand how data moves across it. The Networking Fundamentals learning path
ViewPreview · 30 min
Master security-focused promptfoo evaluations. Learn to layer deterministic regex and semantic LLM-rubric assertions to block RAG context poisoning and PII leaks.
Functional testing checks that the AI application works when a user asks a legitimate question. Security-focused evaluation checks that it survives when someone asks a harmful one.
promptfoo is good at checking valid JSON or a polite tone. In AI security its real strength is running targeted, automated evaluations for known risks. You combine deterministic regex checks with semantic LLM-as-a-Judge scoring. Then you can verify that the model refuses to create malware, blocks PII leakage and resists indirect prompt injection hidden in retrieved context.
For a security analyst this is how you prove that a fixed weakness stays fixed, and how you measure with numbers how well a RAG pipeline resists context poisoning.
All scenarios, configs and targets in this lesson are fictional and made for isolated training labs. Risky inputs and poisoned documents appear only as placeholders.
Create a free account to unlock the rest of this task, complete tasks, and earn Rep.
Sign in