Prompting, Context & Skills
A Complete Guide for Testers and Test Automation Engineers
What This Guide Teaches
This guide teaches you three things that every tester working with AI needs to know:
Prompt Engineering
How to write instructions that get useful, specific output from an AI model every single time. Not sometimes. Every time.
Context Engineering
How to control what the AI knows before it answers. This is the difference between output that is technically correct and output that is correct for your project, your team, and your standards.
Skills (SKILL.md)
How to encode your standards once and have them applied automatically across every session, every tool, and every engineer on your team. Zero setup per session. Consistent output forever.
Who This Guide Is For
Manual Testers
Who have tried AI tools and gotten inconsistent results. You want to understand why and fix it permanently. You do not need to write code. Everything in your reading path is written in plain English.
Automation Engineers
Who write Playwright, Cypress, Selenium, or similar frameworks. You want AI to produce code that matches your project's conventions without re-explaining them every session.
QA Leads and Managers
Who want to standardise how their team uses AI tools and make quality standards self-enforcing rather than dependent on individuals.
Reading Paths by Role
Do not feel obligated to read every chapter. Find your role below and follow that path.
Manual Tester Path
You will finish with a complete skill system for your daily testing work — test cases, bug reports, story reviews, exploratory charters — all produced in your team's format without explaining it every session.
Automation Engineer Path
You will finish with a complete skill system for your automation project — page objects, spec files, fixtures, API tests, debugging — all produced to your project's standards automatically.