embedded-browser-operator

Automates webpage navigation/interaction and extracts structured JSON, screenshots, and PDFs.

9|2|Updated Apr 24, 2026
One-click install
npx skills add https://github.com/TasiTech/tasi-harness --skill embedded-browser-operator
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: embedded-browser-operator
Source: https://github.com/TasiTech/tasi-harness/tree/main/resources/skills/browser/embedded-browser-operator
Command: npx skills add https://github.com/TasiTech/tasi-harness --skill embedded-browser-operator

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It removes the guesswork and brittleness of manual browsing by letting you programmatically open webpages, verify the UI state, interact with elements, and extract reliable content for downstream tasks.

Core Features & Use Cases

  • Web navigation & state verification: Open URLs, confirm current page state, wait for load/conditions, and snapshot the accessibility/semantic tree to anchor actions.
  • Element interaction & automation: Find elements by role/text/label/attributes (or by @e refs), then click, type, select, hover, check, press keys, scroll, and resize viewport for responsive checks.
  • Evidence-grade extraction & debugging: Extract page or selector content as text/HTML/structured JSON, capture screenshots/PDF, and troubleshoot with storage, cookies, console, and network data.

Quick Start

Ask to open a specific webpage, snapshot it, find a target element by visible text or role, extract the relevant section as JSON, and save a screenshot to the workspace.

Frequently Asked Questions about embedded-browser-operator

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

FAQPage Schema
How do I extract structured JSON from a specific webpage selector?

To extract structured JSON from a webpage selector, you open the URL, snapshot the accessibility tree to anchor actions, locate the target element by role or text, and extract its content as structured JSON. This workflow ensures reliable content extraction.

Can I automate browser interactions like clicking and typing for SPA debugging?

Yes, you can automate browser interactions for SPA debugging by finding elements using role, text, or attributes, then executing actions like click, type, select, and hover. You can also inspect console and network data to troubleshoot behavior.

What is the best way to capture evidence-grade screenshots during web extraction?

The best way to capture evidence-grade screenshots during web extraction is to open the webpage, wait for load conditions, snapshot the semantic state, and capture screenshots or PDFs. This approach verifies the UI state before saving the visual evidence.

How do I inspect the accessibility tree to validate page controls?

You inspect the accessibility tree to validate page controls by opening the URL, waiting for the page to load, and taking a snapshot of the semantic tree. This allows you to anchor actions and verify the current state of UI elements.

Does browser automation work with cookies and network data for troubleshooting?

Yes, browser automation works with cookies and network data for troubleshooting. You can optionally inspect console logs, network requests, storage, and cookie data to ensure reliable outcomes during webpage extraction and UI interaction validation.

Why does web extraction fail when page elements are not fully loaded?

Web extraction fails when elements are not fully loaded because the accessibility tree has not stabilized. Using robust waits and scrolling ensures the page state is verified before locating elements and extracting content, preventing brittleness.