Skip to content

README.md

Latest commit

 

History

History
144 lines (91 loc) · 3.67 KB

README.md

File metadata and controls

144 lines (91 loc) · 3.67 KB
Raw

playwright-manual-to-test-generator-info

Deterministic QA tool: manual steps → consistent Playwright TypeScript tests (overview repo)

Playwright Manual-to-Test Generator (Overview)

Manual test steps → deterministic Playwright TypeScript tests

This repository is an overview and documentation landing page for a QA tool built by Ryan Goede (TekQA Consulting) to solve a real-world manual-to-automation migration problem.

👉 Main Repository (Code + Extension)
https://github.com/rmgoede/playwright-manual-to-test-generator

Use this to install and run the generator.

This repo exists to explain what the tool does, why it exists, and how it’s used.

The Problem This Tool Solves

Many teams still maintain large regression suites as:

  • Excel spreadsheets
  • ALM / QC test cases
  • Jira tickets
  • Plain-text procedural steps

Manually converting these into Playwright tests is:

  • Slow
  • Error-prone
  • Inconsistent
  • Hard to scale

Prompting an LLM helps — but direct code generation can vary run to run.

What This Tool Does

The Playwright Manual-to-Test Generator converts plain-text manual test steps into:

  • A single, runnable Playwright TypeScript test
  • Using deterministic generation rules
  • With Playwright-recommended structure and locators
  • No heuristic selector inference, scraping, or DOM inspection
  • Uses a structured JSON contract (V1.5) to enforce consistent output

Output is designed to be:

  • Predictable
  • Readable
  • Easy to refine once the real DOM is inspected

How It Works (Architecture)

The generator supports two modes:

V1 (Direct Generation)
Manual Steps
→ LLM
→ Playwright TypeScript test

V1.5 (Structured Mode)
Manual Steps
→ LLM
→ Structured JSON (schema-enforced)
→ Deterministic Renderer
→ Playwright TypeScript test

Why this matters

Instead of letting the LLM generate final code directly,
V1.5 introduces a structured contract (JSON) between interpretation and execution.

This results in:

  • More consistent output
  • Reduced variability between runs
  • Clear validation boundaries
  • Deterministic final code generation

Key Characteristics

  • Controlled, consistent output (schema + renderer reduce variability)
  • Rule-based selector generation (role/text-based)
  • Produces a clean first draft, not a “magic” passing test
  • Designed for real migration work, not demos

Some selector refinement is expected by design.

How It’s Used

This tool is used:

  • By QA engineers modernizing legacy test suites
  • During manual → Playwright migration efforts
  • As a repeatable workflow for generating first-draft automation
  • As a practical example of build-when-gaps-exist engineering

It reflects a build tooling when needed, not just use tooling mindset.

Screenshots / Demo

Demo GIF below reflects real output from the tool against common automation flows (login, add-to-cart, verification).

Playwright Manual-to-Test Generator demo

Source Code & Quick Start

The generator is fully open-source:

👉 https://github.com/rmgoede/playwright-manual-to-test-generator

Installation and usage instructions are in the main repository README.

Related Work

About

Built by Ryan Goede
Senior QA Engineer | Automation & CI/CD
28+ years across healthcare, banking, government, and e-commerce

This repo documents real tooling built for real automation problems.