Dev Browser

•
1.3k
•

Automates browser interactions, web scraping, and application testing while maintaining persistent page state across executions.

This skill empowers Claude to control a persistent Chromium browser instance, enabling complex web automation and data extraction tasks. Unlike standard automation scripts that start fresh each time, this skill maintains cookies, localStorage, and page state between executions, allowing for incremental workflows like multi-step form filling or navigating authenticated portals. It includes specialized tools for element discovery via ARIA snapshots, making it highly effective at interacting with unknown or complex web layouts while providing visual feedback through screenshots.

主要功能

01Persistent browser sessions that preserve cookies, localStorage, and DOM state across multiple runs.
02ARIA-based element discovery using semantic snapshots for reliable interaction with complex web layouts.
03Support for incremental script execution to build and debug automation workflows step-by-step.
04Visual debugging capabilities including full-page screenshots and element-specific captures.
05Automated environment management for Chromium, Playwright dependencies, and browser profiles.
061,271 GitHub stars

使用场景

01Automating multi-step workflows such as logging into a service and performing complex administrative tasks.
02Scraping data from dynamic, JavaScript-heavy websites that require session persistence.
03Performing end-to-end testing of web applications by simulating real user interactions and verifying UI state.
name dev-browser
description Browser automation with persistent named pages via the dev-browser CLI. Use when users ask to navigate websites, fill forms, take screenshots, extract web data, test web apps, log into sites, or automate browser workflows. Trigger phrases include "go to [url]", "click on", "fill out the form", "take a screenshot", "scrape", "automate", "test the website", "log into", "open the browser", or any browser interaction request.

dev-browser

CLI for controlling a real Chrome with short Puppeteer scripts. One warm daemon; named pages persist between runs.

npm install -g dev-browser   # bun add -g dev-browser works too; the first run downloads the binary if the install script was blocked
dev-browser install          # only if the first run says "No Chrome found"

Run dev-browser --help (full guide; dev-browser help <topic> for one section) before non-trivial work. Quick start:

dev-browser <<'EOF'
const page = await browser.getPage("main");        // named page persists across runs
await page.goto("https://example.com");            // default waitUntil: domcontentloaded
await page.snapshot({ interactive: true })         // ARIA tree with refs; last expression is printed
EOF
dev-browser -e 'const p = await browser.getPage("main"); await p.click("ref/e6"); await p.waitForLoad(); p.url()'

Gotchas: end lines with semicolons (a line starting with ( continues the previous one); return an object as ({ a, b }); page.click never waits (use waitForSelector first); page names are per browser (--headless and headed are separate Chromes and profiles); refs reset on navigation — re-snapshot; file paths resolve against your cwd (uploadFile, screenshot/pdf path); do not run parallel dev-browser calls against the same named page.