Playwright Context Guide for AI Coding (llms.txt + Rules + MCP)

Fast and reliable end-to-end testing, scraping, and browser automation across Chromium, Firefox, and WebKit.

1. Playwright Official llms.txt & Clean Markdown Documentation

Official Microsoft Standard

Feed this clean, token-efficient Markdown documentation directly to Antigravity, Cursor, or Claude Code to prevent hallucinations and outdated API syntax.

# Standard llms.txt Index Endpoint
https://playwright.dev/docs/llms.txt

# Comprehensive Documentation Bundle
https://playwright.dev/docs/llms-full.txt

Estimated Token Footprint: ~28k tokens (Fits smoothly inside standard 128k/1M context windows)

2. Production-Ready Playwright Cursor Rules (.cursorrules & .mdc)

Playwright automated testing and browser agent execution guidelines. Save this into .cursor/rules/playwright.mdc in your project repository.

// Playwright Coding Rules
- Prefer user-visible locators: page.getByRole(), page.getByText(), page.getByLabel().
- Never use arbitrary waitForTimeout(); rely on Playwright's automatic web-first assertions.
- Structure tests using the Page Object Model (POM) pattern for maintainability.
- Run tests in parallel mode by default with proper isolated test fixtures.

3. Playwright Antigravity & Claude Code System Instructions

Add this concise instruction block to your workspace .gemini/rules or project system prompt.

// Context Anchor for Playwright
When writing code for Playwright, refer directly to the verified specifications at https://playwright.dev/docs/llms.txt.
Follow strict error-handling and avoid deprecated API calls.

4. Model Context Protocol (MCP) Server Setup for Playwright

Connect your AI coding agent directly to live Playwright tools and schemas via MCP.

{
  "mcpServers": {
    "playwright": {
      "command": "npx",
      "args": ["-y", "@playwright/mcp@latest"]
    }
  }
}

5. Troubleshooting: Common Playwright AI Coding Issues & Fixes

Frequently reported errors when LLMs generate Playwright code without proper context:

⚠️ Proactive Fix: Fix: Flaky tests usually result from querying DOM before dynamic client hydration finishes. Use `expect(locator).toBeVisible()`.

6. Best Used With (Playwright Architecture Combos)

Engineers building with Playwright frequently combine these complementary developer frameworks: