skills.rest
marktantongco
marktantongcoCOMMUNITY
Explore all skills by this author →
GitHub

Agent Browser

Automate browser navigation and DOM interactions via a headless CLI.

2|Updated Mar 31, 2026
Search Tags:#cli#ui-testing#headless#web-scraping#browser-automation#form-filling
One-click install
npx skills add https://github.com/marktantongco/promptc-os --skill agent-browser-marktantongco
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: Agent Browser
Source: https://github.com/marktantongco/promptc-os/tree/main/skills/agent-browser
Command: npx skills add https://github.com/marktantongco/promptc-os --skill agent-browser-marktantongco

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Agent Browser removes repetitive manual browser interactions by providing a CLI-driven headless browser that agents and scripts can use to navigate pages, interact with elements, capture snapshots, and record sessions.

Core Features & Use Cases

  • Headless CLI Automation: Navigate pages, reload, go back/forward, and close sessions programmatically for automation and testing.
  • Structured Interaction & Snapshotting: Capture accessibility trees and receive stable element refs or use semantic locators to click, fill, type, and press keys reliably.
  • Testing, Scraping & Auth Flows: Save and load session state for authenticated workflows, perform UI testing, scrape structured data, and record video or traces for debugging.
  • Advanced Controls: Route and mock network requests, manage cookies/storage, control viewport and device emulation, and connect via CDP for deeper debugging.

Quick Start

Open the target URL, snapshot interactive elements, fill the form fields by their element refs, and click the submit button to complete the form submission.

Frequently Asked Questions about Agent Browser

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

FAQPage Schema
How do I automate browser tasks from the command line?▼

Automate browser tasks from the command line using a CLI-driven headless browser that navigates pages, captures accessibility tree snapshots, interacts with elements via stable refs, and outputs JSON. It supports programmatic web scraping, form filling, and UI testing without manual interaction.

How does headless browser automation handle form filling and element interaction?▼

Headless browser automation handles form filling by capturing accessibility tree snapshots to generate stable element refs and semantic locators. Scripts use these refs to reliably click, fill, type, and press keys on targeted DOM elements without relying on brittle CSS selectors.

Can I save and load session state for authenticated web scraping workflows?▼

You can save and load session state for authenticated web scraping workflows. The CLI manages cookies and storage, allowing you to persist authenticated sessions across runs, which is essential for scraping structured data behind login walls or automating complex auth flows.

Does this CLI browser automation tool require Node.js and npm to run?▼

This CLI browser automation tool requires Node.js and npm for both installation and runtime support, and it includes a Node.js fallback. Ensure your local environment or CI pipeline has these dependencies properly configured before executing automation scripts.

What is the best way to record and debug automated UI testing sessions?▼

The best way to record and debug automated UI testing sessions is using the CLI's built-in recording features to capture video and traces. You can also connect via Chrome DevTools Protocol (CDP) for deeper debugging and route network requests for comprehensive testing.

When should I not use a headless CLI browser for web scraping?▼

You should not use a headless CLI browser for web scraping when your target pages require complex real-time visual rendering analysis or lack accessible DOM structures. Additionally, environments without Node.js and npm runtime support cannot execute these automation scripts.

Related Skills

Explore More Software Engineering Skills→
AshesOfTheUndeadAshesOfTheUndead

Agent Browser

Automate headless browser navigation, element manipulation, and data extraction via CLI.

Community
Advanced
archibatearchibate
6

slack

Automate Slack web workspace interactions to read messages and extract channel metadata.

Community
Intermediate
MathiasPaulenkoMathiasPaulenko

WaveXis Session Recording

Capture and replay browser interactions as YAML sessions via the WaveXis CLI.

Community
Advanced
JohannesKoningsJohannesKonings

agent-browser

Automate browser navigation, form submission, data extraction, and visual regression testing.

Community
Advanced