dev-browser

Automate browser tasks with persistent page state across sessions.

122|16|Updated Jan 19, 2026
One-click install
npx skills add https://github.com/w-winter/dot314 --skill dev-browser-w-winter
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: dev-browser
Source: https://github.com/w-winter/dot314/tree/main/skills/dev-browser
Command: npx skills add https://github.com/w-winter/dot314 --skill dev-browser-w-winter

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

Browser automation often loses page state between interactions, causing flaky flows across pages and sessions. This Skill preserves cookies, page state, and session data to enable reliable, multi-step browser tasks.

Core Features & Use Cases

  • Two interfaces: CLI for token-efficient commands and Script-based access to full Playwright API.
  • Persistent state: Maintain page state across interactions, sessions, and navigations.
  • Use Case: Automate logging in to a site, filling multi-page forms, taking screenshots for QA, and extracting web data with scripted workflows.

Quick Start

  1. Install the dev-browser Chrome extension from the release page.
  2. Start the relay in Terminal 1: cd ~/.pi/agent/skills/dev-browser && npm run start-extension Wait for "Extension connected".
  3. In Terminal 2, connect via the CLI: devbrowse go "https://example.com/login" devbrowse read devbrowse snap

Frequently Asked Questions about dev-browser

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

FAQPage Schema
How do I keep browser state persistent across multiple Playwright automation sessions?

Persistent browser state is maintained by preserving cookies, page state, and session data across interactions. This prevents flaky flows and enables reliable multi-step web automation tasks like navigating websites and filling multi-page forms.

How do I automate logging into a website and taking screenshots for QA testing?

Automate logging in by using CLI commands to navigate to URLs and read page content, or use scripts for full API access. You can fill multi-page forms and take screenshots for QA while maintaining persistent session data across navigations.

Do I need Node.js and npm to run browser automation with persistent state?

Yes, you need a Node.js runtime with npm to run browser automation with persistent state. It relies on Playwright-based server components and requires starting a local relay service before connecting via the CLI.

Can I use a Chrome extension for web testing instead of the CLI?

Yes, an optional Chrome extension provides an extension mode for web testing. You install the extension, start the relay service in a terminal, and wait for the connection confirmation before executing browser commands.

What's the difference between using CLI commands and scripts for browser automation?

CLI commands offer token-efficient interactions for simple tasks like reading pages and taking screenshots. Scripts provide access to the full Playwright API, enabling complex data extraction and automated web app testing workflows.

Why does my browser automation lose page state and cause flaky flows across pages?

Browser automation loses page state between interactions when session data and cookies are not preserved. Using a persistent state approach maintains this data across navigations and sessions, preventing flaky multi-step flows.