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
Network Forensics
Network traffic is the nervous system of any organization, and for attackers, it is the highway used for initial access,
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
ViewSOC Tools: Hands-On Practicals
Theory is useless without muscle memory. You can read a hundred books on how a SIEM works, but until you’ve stared down
ViewRecon & Enumeration Tools
Reconnaissance is the foundation of every successful penetration test. Amateurs treat it as a box-ticking exercise, runn
ViewExploitation Fundamentals
Amateurs stop at the proof of concept. They pop an alert(1), dump a database via SQLi, or find an IDOR, write a report,
ViewPreview · 30 min
Master promptfoo for LLM regression testing. Learn to detect silent degradation, configure YAML assertions, and integrate automated safety gates into CI/CD.
In normal software, regression testing is basic practice. You change a line of code, then you run automated tests to check that old features still work.
In AI engineering this is even more important, and harder. Large Language Models (LLMs) are not deterministic. A small change can quietly weaken the safety rules or break the output format. Examples: a tweak to the system prompt for a friendlier tone, an upgrade from one model version to a newer one, or a change of the temperature setting.
This is where promptfoo helps. Other tools have other jobs. Garak is built for broad vulnerability discovery. PyRIT is strong at complex, multi-turn adversarial orchestration. promptfoo is the standard open-source command line tool for repeatable evaluation and regression testing. You define a dataset of inputs and the expected behaviour (assertions). promptfoo then checks, every time, that the model meets them.
For a security analyst, promptfoo is the gatekeeper. It makes sure a prompt update does not reopen a weakness that was fixed last week.
All scenarios, configs and targets in this lesson are fictional and made for isolated training labs. Test prompts appear only as placeholders. You test only systems you are authorised to test.
Create a free account to unlock the rest of this task, complete tasks, and earn Rep.
Sign in