dev-browser

Automate browser interactions with persistent page state across scripts.

Updated Nov 17, 2025
One-click install
npx skills add https://github.com/ci/persops --skill dev-browser-ci
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: dev-browser
Source: https://github.com/ci/persops/tree/main/modules/ai/skills/dev-browser
Command: npx skills add https://github.com/ci/persops --skill dev-browser-ci

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

Browser automation with persistent page state addresses the tedium of manual web interactions across repeated tasks. It enables navigating to URLs, filling forms, taking screenshots, scraping data, and running web app tests without losing context between script executions.

Core Features & Use Cases

  • Persistent pages across runs to support multi-step workflows.
  • Standalone mode (local automation) and extension mode (connected browser sessions).
  • Small, focused scripts (one action per script) that can be composed into larger end-to-end tasks.
  • Visual verification via screenshots and straightforward data extraction.
  • Use case: automate login flows, form submissions, and data scraping across multiple sites with state preserved.

Quick Start

Start the standalone server with: ./server.sh & Then add and run small scripts in the scripts/ directory to automate a sample task, or enable extension mode by starting the extension relay and connecting your browser session per the repo README.

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 tasks while keeping page state between script runs?

Browser automation with persistent page state is achieved by running small, focused Playwright scripts through a standalone server or extension relay, preserving navigation context, cookies, and session data across multiple executions.

Can I use Playwright for web scraping and taking screenshots of web apps?

Yes, Playwright supports web scraping and taking screenshots for visual verification. You create deterministic scripts in a scripts/ directory to navigate URLs, extract data, and capture page images without manual browser interaction.

Do I need Node.js and Chromium installed to run browser automation scripts?

Yes, this browser automation approach requires a Node.js environment with Playwright and Chromium installed to execute scripts, navigate sites, fill forms, and maintain persistent page state across runs.

What is the best way to compose large end-to-end web testing workflows?

The best way to compose end-to-end web testing workflows is creating one small action per script in a scripts/ directory, then chaining them together to handle multi-step tasks like login flows and form submissions with state preserved.

Does extension mode let me connect browser automation to my existing browser session?

Yes, extension mode connects browser automation to your existing browser session by starting an extension relay, allowing scripts to interact with live pages rather than launching a standalone local Chromium instance.

Why does my Playwright script lose login state when running multiple form submissions?

Login state is lost without persistent page state. Using a standalone server with persistent pages preserves cookies and session context, ensuring subsequent scripts maintain authentication across multiple form submissions and navigations.