browser

Automate browser interactions and capture compact AI-optimized snapshots of web pages.

1|Updated Jun 9, 2020
One-click install
npx skills add https://github.com/dalager/jernkorsetbreve --skill browser-dalager
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: browser
Source: https://github.com/dalager/jernkorsetbreve/tree/main/.claude/skills/browser
Command: npx skills add https://github.com/dalager/jernkorsetbreve --skill browser-dalager

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill eliminates brittle, verbose DOM scraping and manual browser scripting by providing compact AI-optimized snapshots and stable element references that let agents interact with pages deterministically.

Core Features & Use Cases

  • AI-Optimized Snapshots: Capture accessibility-style snapshots that reference elements (e.g., @e1) to reduce context size and improve reliability.
  • Deterministic Interactions: Click, fill, type, press, hover, select, check/uncheck, and scroll using refs or selectors with wait and load-state controls.
  • Session & State Management: Isolated sessions, state save/load, and multi-session workflows for authenticated scraping, parallel scraping swarms, and replayable login flows.
  • Use Case: Automate login + data extraction from a site by opening the page, snapshotting interactive elements, filling credentials using refs, clicking submit, and waiting for the dashboard to load.

Quick Start

Open a page, take an interactive snapshot, and use the returned element references to fill inputs and click buttons to complete the automation task.

Frequently Asked Questions about browser

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

FAQPage Schema
How do I automate web scraping without brittle DOM selectors?

Automate web scraping using AI-optimized snapshots that capture accessibility-style elements with stable references like @e1. This approach reduces DOM context size and enables deterministic agent-driven navigation, eliminating verbose manual browser scripting.

Can I manage multiple authenticated sessions for parallel scraping?

Yes, you can manage multiple authenticated sessions using session isolation and state save/load features. This enables parallel scraping swarms and replayable login flows by maintaining separate session states for each concurrent workflow.

How do I fill out and submit forms automatically with a browser agent?

Fill and submit forms by taking an interactive snapshot of the page, using the returned element references to fill input fields, and clicking submit buttons. The agent waits for load states to ensure deterministic form-filling automation.

What is the best way to capture compact web page states for AI agents?

Capture compact web page states by generating AI-optimized snapshots that reference interactive elements with stable identifiers. This method reduces context size compared to raw DOM scraping and provides deterministic references for agent interactions.

Does browser automation work for end-to-end UI testing workflows?

Yes, browser automation supports end-to-end UI testing through deterministic interactions like clicking, typing, hovering, and scrolling. It provides wait and load-state controls alongside screenshotting capabilities to verify application behavior reliably.

Why does my web automation script fail when page elements load dynamically?

Dynamic page elements cause failures when scripts lack proper wait and load-state controls. Using AI-optimized snapshots with stable element references and explicit waiting mechanisms ensures agents interact with elements only after they fully load.