Built With
Specwright is built on the shoulders of excellent open-source projects. This page lists the core libraries that power the framework and the documentation site.
E2E Framework (Plugin)
These packages are installed in your project when you run npx @specwright/plugin init:
| Package | Version | Role |
|---|---|---|
@playwright/test | ^1.59.1 | Browser automation engine — runs your tests across Chromium, Firefox, and WebKit |
playwright-bdd | ^8.5.0 | Gherkin BDD layer on top of Playwright — compiles .feature files into Playwright specs via bddgen |
@faker-js/faker | ^10.3.0 | Realistic test data generation — names, emails, phone numbers, addresses, and more |
multiple-cucumber-html-reporter | ^3.10.0 | Generates the BDD HTML report from Cucumber JSON output |
dotenv | ^17.2.0 | Loads .env.testing into process.env at test startup |
AI Pipeline
The agent pipeline that generates and heals tests is powered by:
| Tool | Role |
|---|---|
| Claude (Anthropic) | The AI model driving all agents — exploration, generation, healing, and review |
| Claude Code | CLI and SDK that runs the multi-agent pipeline via skills and agents |
claude-runner | Lightweight SDK used by the Desktop app to spawn and stream Claude agents programmatically |
@playwright/mcp | MCP server that gives agents live browser access during exploration |
Documentation Site
The docs you're reading are built with:
| Package | Role |
|---|---|
| Next.js | React framework powering the docs site |
| Tailwind CSS | Utility-first CSS framework |
| Framer Motion | Animations and transitions |
| MDX | Markdown with React components — how these pages are written |
Acknowledgements
A huge thank you to the maintainers of playwright-bdd — the Gherkin-to-Playwright bridge is what makes the BDD experience in Specwright possible. If you find it useful, consider sponsoring the project.
And to the Playwright team at Microsoft — reliable, fast, and cross-browser test automation out of the box.