dev-browser

Automate browser interactions with a persistent Playwright-based Chromium context.

1|Updated Oct 29, 2025
One-click install
npx skills add https://github.com/thanhtoan105/Accounting_ERP_Chatbot --skill dev-browser-thanhtoan105
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: dev-browser
Source: https://github.com/thanhtoan105/Accounting_ERP_Chatbot/tree/main/.claude/skills/dev-browser
Command: npx skills add https://github.com/thanhtoan105/Accounting_ERP_Chatbot --skill dev-browser-thanhtoan105

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires express, playwright, and includes scripts (resource) components.

What problem does it solve?

Dev Browser provides browser automation with a persistent page state, enabling repeated browser workflows across script executions. It keeps cookies, localStorage, and DOM state inside a running Chromium instance, making multi-step tasks reliable and faster to iterate.

Core Features & Use Cases

  • Persistent Chromium context across runs for reliable multi-step flows
  • REST API page management to get or create named pages
  • Small, focused scripts for deterministic tasks; reuse for repeated work
  • ARIA snapshot tooling to discover elements and persist references
  • Screenshots for verification and rapid debugging
  • CLI-friendly server compatibility with Playwright for real-world automation
  • Use cases include login automation, multi-step form completion, and web data extraction with preserved session state.

Quick Start

Start the dev-browser server:

  • ./skills/dev-browser/server.sh & Then create or fetch a page:
  • curl -X POST http://localhost:9222/pages -H "Content-Type: application/json" -d '{"name":"home"}' Finally, run a scripted action against the page, e.g.:
  • bun x tsx scripts/your_script.ts

Frequently Asked Questions about dev-browser

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate browser interactions while keeping page state between script runs?

Persistent browser automation maintains cookies, localStorage, and DOM state across script executions within a running Chromium instance. Dev Browser provides a REST API to manage named pages, enabling multi-step workflows like login automation and form completion without re-initializing state each time.

Can I use Playwright to control a browser with a persistent context across multiple automation tasks?

Yes. Dev Browser wraps Playwright with a persistent Chromium context accessible via REST API, allowing named pages to retain session state. Start the server, create or fetch pages by name, and run scripts against them while preserving cookies and DOM between executions.

How do I discover elements and automate clicking, form filling, and screenshots in web apps?

Dev Browser includes ARIA snapshot tooling to locate elements deterministically, REST endpoints to execute inline scripts, and screenshot capture for verification. Use these to automate clicking, filling forms, scraping data, and testing workflows with persisted session context.

What's the fastest way to run repeated web automation workflows without rebuilding browser state each time?

Persistent browser automation caches state inside a long-running Chromium instance, eliminating initialization overhead. Dev Browser manages named pages through a CLI-friendly server, letting you write small, focused scripts that reuse the same page context and stored credentials for rapid iteration.

Can I automate web scraping and data extraction while maintaining login sessions across multiple pages?

Dev Browser preserves cookies and localStorage in a persistent Chromium context, enabling web scraping and data extraction to run across multiple pages without re-authenticating. Use the REST API to manage named pages and run scripts that extract data while maintaining session state.