skills.rest
ajbmachon
ajbmachonCOMMUNITY
Explore all skills by this author →
Website𝕏@ajbmachon2GitHub

dev-browser

Automate browser navigation and interaction with persistent named page state.

7|2|Updated Nov 30, 2025
Search Tags:#browser automation#web scraping#form filling#playwright#extension mode#aria snapshot#persistent sessions
One-click install
npx skills add https://github.com/ajbmachon/ajbm-skills --skill dev-browser-ajbmachon
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: dev-browser
Source: https://github.com/ajbmachon/ajbm-skills/tree/main/.claude/skills/dev-browser
Command: npx skills add https://github.com/ajbmachon/ajbm-skills --skill dev-browser-ajbmachon

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires playwright, express, hono, tsx, @hono/node-server, @hono/node-ws, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill removes the friction of manual website interaction by letting you automate navigation, form filling, screenshots, and data extraction while keeping browser state (cookies, logins, page context) across steps.

Core Features & Use Cases

  • Persistent page state: Keeps named pages alive on the server so you can build workflows incrementally.
  • Standalone and extension-backed modes: Either launch a fresh Chromium session or control the user’s existing logged-in Chrome via the companion extension.
  • ARIA snapshot element discovery: Uses accessibility-tree snapshots and stable refs to reliably click and fill even on complex/unknown layouts.
  • Scraping via network replay: Supports capturing and replaying API calls for faster, more reliable large-scale extraction than scrolling the DOM.
  • Debug-friendly execution loop: Recommends small scripts, state inspection (screenshots + snapshots), and iterative refinement with error recovery.

Quick Start

Launch the server in standalone mode, then tell the Skill what URL to open and which action to perform next.

Frequently Asked Questions about dev-browser

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

FAQPage Schema
How do I automate web scraping with persistent login sessions across multiple steps?▼

Automate web scraping with persistent login sessions by using server-backed Playwright scripting that keeps named pages alive on the server, preserving cookies and page context across multi-step workflows. This allows you to build workflows incrementally without re-authenticating.

What is ARIA snapshot element discovery for browser automation?▼

ARIA snapshot element discovery for browser automation uses accessibility-tree snapshots and stable refs to reliably click and fill elements on complex or unknown layouts. The getAISnapshot and selectSnapshotRef functions provide deterministic element targeting without fragile DOM selectors.

How do I scrape web data by replaying network requests instead of scrolling the DOM?▼

Scrape web data by replaying network requests through the Skill's network replay feature, which captures and replays API calls for faster, more reliable large-scale extraction than scrolling the DOM. This approach bypasses rendering overhead and directly targets backend data endpoints.

Can I use Playwright to control my existing logged-in Chrome browser?▼

You can use Playwright to control your existing logged-in Chrome browser through the companion extension mode. This standalone and extension-backed capability lets you either launch a fresh Chromium session or automate interactions within your already authenticated browser environment.

Does dev-browser support taking screenshots and filling forms during multi-step flows?▼

dev-browser supports taking screenshots and filling forms during multi-step flows through its persistent named page state and debug-friendly execution loop. It recommends small scripts, state inspection via screenshots and snapshots, and iterative refinement with error recovery for reliable automation.

What are the limitations of using extension mode for website automation?▼

The limitations of using extension mode for website automation include dependency on the companion Chrome extension and the requirement of a root SKILL.md entry with YAML frontmatter. Deterministic execution depends on server-backed Playwright scripting using ARIA snapshots for reliable element targeting.

Related Skills

Explore More Software Engineering Skills→
cshyangcshyang

trigger-config

Configure Trigger.dev projects with trigger.config.ts for build extensions.

Community
Advanced
rar-filerar-file
42

JSON Formatter

Parse, transform, and validate structured data in JSON, YAML, XML, and related formats.

Community
Intermediate
battyejpbattyejp

dev-browser

Automate browser interactions with persistent sessions using Node and Playwright.

Community
Advanced
dglijin-ossdglijin-oss
10

browser-use

Automate browser navigation, form filling, screenshots, and data extraction.

Community
Advanced