skills.rest
MerkyorLynn
MerkyorLynnCOMMUNITY
Explore all skills by this author →
GitHub

Agent Browser

Control a headless browser via deterministic commands and element refs.

37|4|Updated Mar 31, 2026
Search Tags:#data extraction#form filling#ui testing#web automation#headless browser#snapshot refs#accessibility tree
One-click install
npx skills add https://github.com/MerkyorLynn/Lynn --skill agent-browser-merkyorlynn
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: Agent Browser
Source: https://github.com/MerkyorLynn/Lynn/tree/main/skills2set/agent-browser
Command: npx skills add https://github.com/MerkyorLynn/Lynn --skill agent-browser-merkyorlynn

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Agent Browser removes the manual, error-prone work of interacting with websites by letting you drive a headless browser through structured commands and capture page state for reliable follow-up actions.

Core Features & Use Cases

  • Headless browser automation: Navigate, click, fill forms, scroll, drag-and-drop, and handle dialogs consistently.
  • Snapshot-based interaction: Analyze pages via accessibility trees and interact using stable element refs (e.g., @e1, @e2).
  • Data extraction & diagnostics: Get text/HTML/values/attributes, capture screenshots/PDFs, and collect console/errors for debugging.
  • Use Case: Quickly complete a multi-step login and form submission flow, then extract confirmation details (like reference IDs or totals) from the resulting page.

Quick Start

Open a target URL, take an interactive snapshot, then click the correct element by its ref to drive the workflow end to end.

Frequently Asked Questions about Agent Browser

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

FAQPage Schema
How do I automate web form filling and data extraction without scripts breaking on DOM changes?▼

Automate web form filling and data extraction by controlling a headless browser to take interactive snapshots of the accessibility tree, then click or fill elements using stable refs like @e1. If the DOM changes, re-snapshotting provides new refs for subsequent actions.

What is snapshot-based interaction for headless browser automation?▼

Snapshot-based interaction in headless browser automation analyzes pages via accessibility trees to generate stable element references (e.g., @e1, @e2). You use these deterministic refs to reliably click, fill, and extract data without relying on brittle CSS selectors.

Can I use web automation to capture screenshots and console errors for UI regression testing?▼

Yes, you can use this web automation approach to capture screenshots, generate PDFs, and collect console errors during UI regression testing. Deterministic command-driven interactions navigate dynamic pages and capture diagnostic output for debugging workflows.

How do I extract structured data from dynamic web pages using a headless browser?▼

Extract structured data from dynamic web pages by navigating to the target URL, taking an interactive snapshot, and executing commands to get text, HTML, values, or attributes. Optional JSON output formats the extracted data for machine-readable consumption.

Does headless browser automation support waiting conditions and drag-and-drop interactions?▼

Headless browser automation supports waiting conditions and drag-and-drop interactions alongside navigation, scrolling, and dialog handling. These deterministic command-driven actions ensure consistent execution of complex multi-step workflows like login and submission flows.

Agent Browser requires no external dependencies, but what environment is needed to run it?▼

Agent Browser runs without external dependencies, requiring only an environment capable of executing a headless browser. It operates through structured commands to navigate, snapshot, and interact with web pages, returning accessibility tree refs and optional JSON output.

Related Skills

Explore More Software Engineering Skills→
liuaibin001liuaibin001

automating-browsers

Automate browser interactions for AI agents using the agent-browser CLI.

Community
Advanced
alibabaalibaba
10.5k

pdf-extractor

Extract text, tables, and metadata from PDF documents into JSON.

Official
Intermediate
proffesor-for-testingproffesor-for-testing
436

contract-testing

Verify microservice API providers against consumer-driven Pact contracts in CI/CD pipelines.

Community
Advanced
MathiasPaulenkoMathiasPaulenko

WaveXis Web Scraping

Automates web scraping and site crawling with browser-based stealth.

Community
Advanced