playwright-skill

Name: playwright-skill
Author: LambdaTest

335stars67forks

How to install playwright-skill?

lambdatest/agent-skills/playwright-skill

$npx -y skills add LambdaTest/agent-skills --skill playwright-skill

Installs into the current project.

Summary

Generates production-grade Playwright automation scripts and E2E tests in TypeScript, JavaScript, Python, Java, or C#. Supports local execution and TestMu AI cloud across 3000+ browser/OS combinations and real mobile devices. Use when the user asks to write Playwright tests, auto

Files · 1

View on GitHub

SKILL.md

1	# Playwright Test Automation
2
3	## Step 1 — Determine Execution Target
4
5	Decide BEFORE writing any code:
6
7	\| User says... \| Target \| Action \|
8	\|---\|---\|---\|
9	\| No cloud mention, "locally", "debug" \| Local \| Standard Playwright config \|
10	\| "cloud", "TestMu", "LambdaTest", "cross-browser", "real device" \| Cloud \| See [reference/cloud-integration.md](reference/cloud-integration.md) \|
11	\| Impossible local combo (Safari on Windows, Edge on Linux) \| Cloud \| Suggest TestMu AI, see [reference/cloud-integration.md](reference/cloud-integration.md) \|
12	\| "HyperExecute", "parallel at scale" \| HyperExecute \| Defer to `hyperexecute-skill` \|
13	\| "visual regression", "screenshot comparison" \| SmartUI \| Defer to `smartui-skill` \|
14	\| Ambiguous \| Local \| Default local, mention cloud option \|
15
16	## Step 2 — Detect Language
17
18	\| Signal \| Language \| Default \|
19	\|---\|---\|---\|
20	\| "TypeScript", "TS", `.ts`, or no language specified \| TypeScript \| ✅ \|
21	\| "JavaScript", "JS", `.js` \| JavaScript \| \|
22	\| "Python", "pytest", `.py` \| Python \| See [reference/python-patterns.md](reference/python-patterns.md) \|
23	\| "Java", "Maven", "Gradle", "TestNG" \| Java \| See [reference/java-patterns.md](reference/java-patterns.md) \|
24	\| "C#", ".NET", "NUnit", "MSTest" \| C# \| See [reference/csharp-patterns.md](reference/csharp-patterns.md) \|
25
26	## Step 3 — Determine Scope
27
28	\| Request type \| Output \|
29	\|---\|---\|
30	\| One-off quick script \| Standalone `.ts` file, no POM \|
31	\| Single test for existing project \| Match their structure and conventions \|
32	\| New test suite / project \| Full scaffold — see [scripts/scaffold-project.sh](scripts/scaffold-project.sh) \|
33	\| Fix flaky test \| Debugging checklist — see [reference/debugging-flaky.md](reference/debugging-flaky.md) \|
34	\| API mocking needed \| See [reference/api-mocking-visual.md](reference/api-mocking-visual.md) \|
35	\| Mobile device testing \| See [reference/mobile-testing.md](reference/mobile-testing.md) \|
36
37	---
38
39	## Core Patterns — TypeScript (Default)
40
41	### Selector Priority
42
43	Use in this order — stop at the first that works:
44
45	1. `getByRole('button', { name: 'Submit' })` — accessible, resilient
46	2. `getByLabel('Email')` — form fields
47	3. `getByPlaceholder('Enter email')` — when label missing
48	4. `getByText('Welcome')` — visible text
49	5. `getByTestId('submit-btn')` — last resort, needs `data-testid`
50
51	Never use raw CSS/XPath unless matching a third-party widget with no other option.
52
53	### Assertions — Always Web-First
54
55	```typescript
56	// ✅ Auto-retries until timeout
57	await expect(page.getByRole('heading')).toBeVisible();
58	await expect(page.getByRole('alert')).toHaveText('Saved');
59	await expect(page).toHaveURL('/dashboard');
60
61	// ❌ No auto-retry — races with DOM
62	const text = await page.textContent('.msg');
63	expect(text).toBe('Saved');
64	```
65
66	### Anti-Patterns
67
68	\| ❌ Don't \| ✅ Do \| Why \|
69	\|----------\|-------\|-----\|
70	\| `page.waitForTimeout(3000)` \| `await expect(locator).toBeVisible()` \| Hard waits are flaky \|
71	\| `expect(await el.isVisible())` \| `await expect(el).toBeVisible()` \| No auto-retry \|
72	\| `page.$('.btn')` \| `page.getByRole('button')` \| Fragile selector \|
73	\| `page.click('.submit')` \| `page.getByRole('button', {name:'Submit'}).click()` \| Not accessible \|
74	\| Shared state between tests \| `test.beforeEach` for setup \| Tests must be independent \|
75	\| `try/catch` around assertions \| Let Playwright handle retries \| Swallows real failures \|
76
77	### Page Object Model
78
79	Use POM for any project with more than 3 tests. Full patterns with base page, fixtures, and examples in [reference/page-object-model.md](reference/page-object-model.md).
80
81	Quick example:
82
83	```typescript
84	// pages/login.page.ts
85	import { Page, Locator } from '@playwright/test';
86
87	export class LoginPage {
88	readonly emailInput: Locator;
89	readonly passwordInput: Locator;
90	readonly submitButton: Locator;
91
92	constructor(private page: Page) {
93	this.emailInput = page.getByLabel('Email');
94	this.passwordInput = page.getByLabel('Password');
95	this.submitButton = page.getByRole('button', { name: 'Sign in' });
96	}
97
98	async login(email: string, password: string) {
99	await this.emailInput.fill(email);
100	await this.passwordInput.fill(password);
101	await this.submitButton.click();
102	}
103	}
104	```
105
106	### Configuration — Local
107
108	```typescript
109	// playwright.conf

People who liked this also viewed

6 picks

Type

playwright-skill

335stars67forks

View on GitHub

How to install playwright-skill?

lambdatest/agent-skills/playwright-skill

$npx -y skills add LambdaTest/agent-skills --skill playwright-skill

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Playwright Test Automation
2
3	## Step 1 — Determine Execution Target
4
5	Decide BEFORE writing any code:
6
7	\| User says... \| Target \| Action \|
8	\|---\|---\|---\|
9	\| No cloud mention, "locally", "debug" \| Local \| Standard Playwright config \|
10	\| "cloud", "TestMu", "LambdaTest", "cross-browser", "real device" \| Cloud \| See [reference/cloud-integration.md](reference/cloud-integration.md) \|
11	\| Impossible local combo (Safari on Windows, Edge on Linux) \| Cloud \| Suggest TestMu AI, see [reference/cloud-integration.md](reference/cloud-integration.md) \|
12	\| "HyperExecute", "parallel at scale" \| HyperExecute \| Defer to `hyperexecute-skill` \|
13	\| "visual regression", "screenshot comparison" \| SmartUI \| Defer to `smartui-skill` \|
14	\| Ambiguous \| Local \| Default local, mention cloud option \|
15
16	## Step 2 — Detect Language
17
18	\| Signal \| Language \| Default \|
19	\|---\|---\|---\|
20	\| "TypeScript", "TS", `.ts`, or no language specified \| TypeScript \| ✅ \|
21	\| "JavaScript", "JS", `.js` \| JavaScript \| \|
22	\| "Python", "pytest", `.py` \| Python \| See [reference/python-patterns.md](reference/python-patterns.md) \|
23	\| "Java", "Maven", "Gradle", "TestNG" \| Java \| See [reference/java-patterns.md](reference/java-patterns.md) \|
24	\| "C#", ".NET", "NUnit", "MSTest" \| C# \| See [reference/csharp-patterns.md](reference/csharp-patterns.md) \|
25
26	## Step 3 — Determine Scope
27
28	\| Request type \| Output \|
29	\|---\|---\|
30	\| One-off quick script \| Standalone `.ts` file, no POM \|
31	\| Single test for existing project \| Match their structure and conventions \|
32	\| New test suite / project \| Full scaffold — see [scripts/scaffold-project.sh](scripts/scaffold-project.sh) \|
33	\| Fix flaky test \| Debugging checklist — see [reference/debugging-flaky.md](reference/debugging-flaky.md) \|
34	\| API mocking needed \| See [reference/api-mocking-visual.md](reference/api-mocking-visual.md) \|
35	\| Mobile device testing \| See [reference/mobile-testing.md](reference/mobile-testing.md) \|
36
37	---
38
39	## Core Patterns — TypeScript (Default)
40
41	### Selector Priority
42
43	Use in this order — stop at the first that works:
44
45	1. `getByRole('button', { name: 'Submit' })` — accessible, resilient
46	2. `getByLabel('Email')` — form fields
47	3. `getByPlaceholder('Enter email')` — when label missing
48	4. `getByText('Welcome')` — visible text
49	5. `getByTestId('submit-btn')` — last resort, needs `data-testid`
50
51	Never use raw CSS/XPath unless matching a third-party widget with no other option.
52
53	### Assertions — Always Web-First
54
55	```typescript
56	// ✅ Auto-retries until timeout
57	await expect(page.getByRole('heading')).toBeVisible();
58	await expect(page.getByRole('alert')).toHaveText('Saved');
59	await expect(page).toHaveURL('/dashboard');
60
61	// ❌ No auto-retry — races with DOM
62	const text = await page.textContent('.msg');
63	expect(text).toBe('Saved');
64	```
65
66	### Anti-Patterns
67
68	\| ❌ Don't \| ✅ Do \| Why \|
69	\|----------\|-------\|-----\|
70	\| `page.waitForTimeout(3000)` \| `await expect(locator).toBeVisible()` \| Hard waits are flaky \|
71	\| `expect(await el.isVisible())` \| `await expect(el).toBeVisible()` \| No auto-retry \|
72	\| `page.$('.btn')` \| `page.getByRole('button')` \| Fragile selector \|
73	\| `page.click('.submit')` \| `page.getByRole('button', {name:'Submit'}).click()` \| Not accessible \|
74	\| Shared state between tests \| `test.beforeEach` for setup \| Tests must be independent \|
75	\| `try/catch` around assertions \| Let Playwright handle retries \| Swallows real failures \|
76
77	### Page Object Model
78
79	Use POM for any project with more than 3 tests. Full patterns with base page, fixtures, and examples in [reference/page-object-model.md](reference/page-object-model.md).
80
81	Quick example:
82
83	```typescript
84	// pages/login.page.ts
85	import { Page, Locator } from '@playwright/test';
86
87	export class LoginPage {
88	readonly emailInput: Locator;
89	readonly passwordInput: Locator;
90	readonly submitButton: Locator;
91
92	constructor(private page: Page) {
93	this.emailInput = page.getByLabel('Email');
94	this.passwordInput = page.getByLabel('Password');
95	this.submitButton = page.getByRole('button', { name: 'Sign in' });
96	}
97
98	async login(email: string, password: string) {
99	await this.emailInput.fill(email);
100	await this.passwordInput.fill(password);
101	await this.submitButton.click();
102	}
103	}
104	```
105
106	### Configuration — Local
107
108	```typescript
109	// playwright.conf

People who liked this also viewed

6 picks

Type