AI Test Generation Cheat Sheet

A quick reference for using AI to generate test plans, scenarios, bug reports, metrics, and automation starter code with QA guardrails.

AI Module 4 Test generation QA workflows

Requirement Analysis

Ask AI to find missing assumptions, risks, and unclear rules before test generation.

Prompt for:
- assumptions
- edge cases
- missing business rules
- clarifying questions

Use templates so AI output follows a predictable structure.

Sections:
- scope
- risks
- environment
- strategy
- entry/exit criteria
- timeline

Always ask for grouped positive, negative, boundary, and security coverage.

Return:
- title
- priority
- preconditions
- steps
- expected result

AI can accelerate strategy drafts, but humans still own tradeoffs and final approval.

Useful for:
- scope breakdown
- risk-based prioritization
- module-level focus
- test type mapping

AI helps turn raw notes into cleaner, more actionable defect reports.

Good bug format:
- summary
- environment
- steps
- actual result
- expected result
- severity
- attachments needed

AI can summarize metrics, trends, and anomalies, but the source numbers must be trusted first.

Useful summaries:
- pass/fail trend
- flaky test hotspots
- coverage gaps
- failure clusters

Generated code is a starting point, not a finished framework artifact.

Ask for:
- locator choices
- assertions
- page object suggestions
- edge case coverage
- maintainability notes

Generated output should be validated before it becomes shared team knowledge or real code.

Review for:
- correctness
- duplicates
- missing risks
- unrealistic steps
- brittle selectors

The highest-value sequence is analyze, generate, review, refine, then implement.

Req -> analysis -> scenarios -> review -> code draft -> human approval