CreatmanCEO

Webtest Orch — Testing skill for Claude Code

Testing community

Token-efficient e2e orchestration skill for Claude Code: explore once, replay deterministically.

How to install Webtest Orch

This entry records only its repository, not the path inside it, so there is no exact command to give. Open CreatmanCEO/webtest-orch and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Webtest Orch does

Token-efficient e2e orchestration skill for Claude Code: explore once, replay deterministically. Playwright + axe-core + run-diff. Tests stay in your repo. MIT.

Alternatives in Testing

Testing Strategy — This project is small, runs in a terminal, and is mostly deterministic 10.9k ★
PR Visual Diff — Post before/after stills + GIFs on a visual PR, captured headlessly with Playwright + ffmpeg 324 ★
Design QA — Visual QA on an HTML/CSS implementation — checks responsive behavior at mobile/tablet/desktop, design token co 309 ★

README

webtest-orch

[![CI](https://github.com/CreatmanCEO/webtest-orch/actions/workflows/ci.yml/badge.svg)](https://github.com/CreatmanCEO/webtest-orch/actions/workflows/ci.yml) [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE) [![npm](https://img.shields.io/npm/v/webtest-orch/beta?label=npm%40beta&color=cb3837)](https://www.npmjs.com/package/webtest-orch) [![Python 3.10+](https://img.shields.io/badge/python-3.10%2B-blue.svg)](https://www.python.org/downloads/) [![Node 18+](https://img.shields.io/badge/node-18%2B-339933.svg)](https://nodejs.org) [![Claude Code](https://img.shields.io/badge/Claude%20Code-Opus%204.7%20%C2%B7%201M%20context-cc785c)](https://code.claude.com)

🇬🇧 English · [🇷🇺 Русский](README.ru.md)

**Token-efficient e2e orchestration skill for Claude Code. Explore once with Playwright MCP — ARIA snapshots, not images. Replay deterministically with `npx playwright test` — ~zero LLM tokens. Bug fingerprinting + run-diff (new / regression / persisting / fixed) out of the box. Tests live in your repo as plain `*.spec.ts`. MIT.**

**Why this exists, in one number:** Playwright MCP burns ~**1.5M tokens** verifying an e-commerce checkout. Playwright **CLI** does the same in ~**25–27k** ([Özal benchmark](https://github.com/microsoft/playwright-mcp/issues/889), [TestDino](https://testdino.com/blog/playwright-cli/), [Morph](https://scrolltest.medium.com/playwright-mcp-burns-114k-tokens-per-test-the-new-cli-uses-27k-heres-when-to-use-each-65dabeaac7a0)). The fix isn't "use less Playwright MCP" — it's split **exploration** (LLM-driven, generates spec.ts) from **replay** (deterministic, runs forever). webtest-orch is the orchestration layer that does both.