Playwright ships the tools.Wrightbench gives them one window.

Wrightbench embeds Playwright's own codegen, UI Mode, Trace Viewer, and reports in one desktop studio — a click each, across every project, with every run kept.

Choose your installer Read the docs

Why Wrightbench

Four tools shouldn't need four windows.

Playwright ships a world-class toolkit — codegen, UI Mode, the Trace Viewer, HTML reports. The CLI scatters it: one command and one window per tool, for one project at a time, and the next run overwrites the last report. Wrightbench keeps the tools and loses the shuffle.

Playwright from the terminal

One command and one window per tool — per project.

Codegen, UI Mode, the report, and traces each launch separately. Switch projects and the shuffle starts over; run again and the last report is overwritten.

zsh — web-app

$ npx playwright codegenwindow 1

$ npx playwright test --uiwindow 2

$ npx playwright show-reportwindow 3

$ npx playwright show-tracewindow 4

$ cd ../admin-portalstart over

# close a window and its results are gone

Playwright in Wrightbench

Every tool in one window — every run kept.

Record, Run, and Report sit side by side in one workspace. Switch projects from the switcher, and every run lands in local history with its report and traces attached.

Wrightbenchweb-app

RecordRunReport

UI Mode · login.spec.tspassed · 1.07s

admin-portal · Playwright 1.61.1one click away

Run #14 saved to historyreports + traces kept

Inside the window

One window, exploded.

Every layer here is Playwright's own tooling — codegen, UI Mode, the Trace Viewer, the HTML report — embedded without recreating an interface. Scroll, and Wrightbench closes the stack into a single window: every project one switcher away, every run kept.

web-appRecordRunReportUI Mode ready

Wrightbench

Workspace shell

One switcher for every project. Record, Run, and Report — a click each.

Recording · Inspector attachedlogin.spec.ts

await page.goto('/login');

await page.getByLabel('Email').fill('mira@example.com');

await page.getByRole('button', { name: 'Sign in' }).click();

await expect(page).toHaveURL('/dashboard');

Playwright · native

Codegen

Playwright’s Inspector, embedded — it writes real Playwright code.

UI Mode · watching3 tests

cart.spec.tspassed · 1.2s

checkout.spec.tsrunning

Playwright · native

UI Mode

Hosted live, without recreating its interface.

Run #14 · HTML reportPassed · 2.6s

OverviewVideoScreenshotsTrace

Before hooks

Navigate

Fill + click

Expect

trace attached · open in Trace Viewer

Playwright · native

Report & Trace Viewer

Playwright’s HTML report and traces, opened in place.

Run #13 · flaky — report + trace kept

Run #12 · passed — report + trace kept

Run #14passedsaved to history8 runs kept · reports + traces attached

Wrightbench

Run history

New runs land on top of old ones — every run kept, report and traces attached.

Scroll to assemble the window

Everything Playwright ships, one window away — for every project, with every run kept.

WrightbenchWorkspace shell

Playwright · nativeCodegen

Playwright · nativeUI Mode

Playwright · nativeReport & Trace Viewer

WrightbenchRun history

Native recording

Record with Playwright’s own Inspector.

Launch the project-local browser and capture flows with the embedded native Inspector—the same codegen the CLI ships, generating real Playwright code. No proprietary recorder to learn.

Embedded UI Mode

Run in Playwright’s native UI Mode.

Wrightbench hosts Playwright UI Mode without recreating its interface. Run, filter, watch, and debug in the real thing—kept alive as you move between workflows.

Projects & versions

Every Playwright project in one window.

Import multiple projects and switch between them without touching a terminal. Wrightbench resolves each project-local configuration and supported Playwright version without rewriting your files.

playwright.config.ts

web-appPlaywright 1.62.1

admin-portalPlaywright 1.61.1

api-e2ePlaywright 1.60.1

Run history

Every run, kept.

The CLI’s HTML report is overwritten by the next run. Wrightbench records every run locally—pass rates, flaky classifications, duration trends, reports, traces, and captured evidence.

Pass rate

14 local runs

78.6%

Run #1Run #14

Run #141 passed2s

Run #131 failed0s

Local-first privacy

No account. No cloud. Your test data stays local.

Wrightbench works directly with local projects and keeps run history, reports, traces, and evidence on your machine—no login or hosted workspace required.

No account or login

No cloud workspace

On-device history

Desktop downloads