Back to Testing & Quality Assurance

Dev Browser

automationplaywrightweb-scrapingbrowser-testingcli
4.3 (203)6.3k📄 MIT🕒 2026-06-05Source ↗

Install this skill

npx skills add SawyerHood/dev-browser

Works across Claude Code, Cursor, Codex, Copilot & Antigravity

What this skill does

  • Executes sandboxed JavaScript automation scripts
  • Maintains persistent browser sessions for multi-step tasks
  • Supports both headless mode and existing Chrome instance attachment
  • Provides full access to Playwright API primitives
  • Captures screenshots and extracts DOM elements programmatically

When to use it

  • Scraping data from sites that require complex state or navigation
  • Testing UI workflows in web applications during development
  • Automating form submission processes across multiple pages
  • Monitoring web content for specific changes or data updates

When not to use it

  • Tasks requiring human-like anti-bot bypass mechanisms
  • High-frequency scraping that requires rotating IP addresses or proxies
  • Extensive browser-based media rendering or heavy graphical tasks

How to invoke it

Example prompts that trigger this skill:

  • Go to the admin dashboard at localhost:3000 and fill out the login form
  • Navigate to the pricing page and take a screenshot of the main table
  • Scrape all product titles from the search results page
  • Test if the registration button remains clickable after a 5-second delay
  • Connect to my running browser and list all currently open tabs

Example workflow

  1. Launch the browser instance using the CLI
  2. Navigate to the target website URL
  3. Execute a script to wait for specific DOM elements to load
  4. Fill in credentials and trigger the login action
  5. Extract the required data into a local JSON object

Prerequisites

  • Node.js installed on the host machine
  • Playwright and Chromium (via dev-browser install command)

Pitfalls & limitations

  • !Sandboxing restricts access to local file system APIs
  • !Browser processes require sufficient memory for headful operations
  • !Initial setup requires manual remote-debugging port configuration for Chrome attachment

FAQ

How does this differ from standard Playwright scripts?
It wraps Playwright in a sandboxed, agent-friendly CLI, simplifying execution for AI agents and CLI-based workflows.
Can I connect this to my current Chrome session?
Yes, by enabling remote debugging at chrome://inspect/#remote-debugging and using the --connect flag.
Is the execution environment secure?
Yes, scripts are executed inside a QuickJS WASM sandbox, ensuring they cannot access the host machine's sensitive files or system variables.

How it compares

Unlike manual browsing or generic prompts, Dev Browser maintains persistent state across distinct script executions, allowing for long-running, reproducible automation workflows.

Source & trust

6.3k stars📄 MIT🕒 Updated 2026-06-05🛡 runs-shell

From the source: “# Dev Browser A CLI for controlling browsers with sandboxed JavaScript scripts. ## Installation ```bash npm install -g dev-browser dev-browser install ``` ## Usage Run `dev-browser --help` to learn more.…”

View the full SKILL.md source

# Dev Browser

A CLI for controlling browsers with sandboxed JavaScript scripts.

## Installation

```bash
npm install -g dev-browser
dev-browser install
```

## Usage

Run `dev-browser --help` to learn more.

Quoted from SawyerHood/dev-browser for reference — see the original for the authoritative, latest version.

📄 Full skill instructions — original source: SawyerHood/dev-browser
Dev Browser is an automation tool that bridges the gap between AI agents and browser environments using sandboxed JavaScript. It provides a CLI interface to control headless browsers or attach to existing Chrome instances for web interactions. By executing scripts within a QuickJS WASM sandbox, it ensures that your browsing actions remain isolated from host system resources. Developers use this skill to automate routine web navigation, handle form submissions, extract structured data, or perform end-to-end testing without manual interaction. Because it maintains persistent page states, you can execute multiple scripts sequentially while keeping your browser session intact. This approach is superior for tasks involving dynamic content, multi-step authentications, or repetitive site monitoring where standard web scrapers often fail due to session resets or complex client-side JavaScript execution requirements.

How to Use This Skill Unit

Option A: Project-Specific (Recommended)

  1. Click "Download" above
  2. In your project, create the directory: .agent/skills/dev-browser/
  3. Save the file as SKILL.md
  4. The agent will automatically discover the skill based on its description.

Option B: Global Installation (All Agents)

Save the file to these locations to make it available across all projects:

  • Claude Code: ~/.claude/skills/SawyerHood/dev-browser/dev-browser/SKILL.md
  • Cursor: ~/.cursor/skills/SawyerHood/dev-browser/dev-browser/SKILL.md
  • Antigravity: ~/.gemini/antigravity/skills/SawyerHood/dev-browser/dev-browser/SKILL.md

🚀 Install with CLI:
npx skills add SawyerHood/dev-browser

Read the Master Guide: Mastering Agent Skills

Recommended Rules

View more rules

Recommended Workflows

View more workflows

Recommended MCP Servers

View more MCP servers

Take It Further

Maximize your productivity with these powerful resources

📋

Define Your Standards

Set up coding standards to ensure this workflow produces consistent, high-quality results.

Browse Rules Library
📖

Master Workflows

Learn how to create custom workflows, use Turbo Mode, and build your automation library.

Complete Guide

How to use this Skill in Claude Code & Cursor

For Claude Code (CLI)

To use this skill in Claude Code, copy the rule content into your project's custom instructions or follow our Add-Skill CLI guide. This ensures Claude follows your standards during every code generation.

For Cursor & Windsurf

For Cursor or Windsurf, individual skills are best used in the "Rules for AI" section. This specific unit helps the agent avoid testing & quality assurance issues, leading to cleaner, more efficient code.

Why the skill format matters: the standardized Agent Skills format lets your AI agent load detailed instructions only when they are relevant, keeping your prompt clean while improving results.

Source & attribution

This skill is categorized under Testing & Quality Assurance and is published by SawyerHood, maintained in SawyerHood/dev-browser.

← Browse All Agent Skills
Sponsored AI assistant. Recommendations may be paid.