agent-browser

Automate browser interactions for testing, form filling, and data extraction.

63|2|Updated Mar 23, 2026
One-click install
npx skills add https://github.com/stolinski/s-stack --skill agent-browser-stolinski
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-browser
Source: https://github.com/stolinski/s-stack/tree/main/skills/agent-browser
Command: npx skills add https://github.com/stolinski/s-stack --skill agent-browser-stolinski

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Automates repetitive browser tasks such as navigating pages, filling forms, taking screenshots, and extracting data, reducing manual work and human error.

Core Features & Use Cases

  • Snapshot-based element references to drive interactions without sending full DOM.
  • Form automation, screenshot capture, and data extraction for testing, QA, and demos.
  • Reference-led workflows that support debugging, documentation, and automated testing.

Quick Start

Open a page with agent-browser open https://example.com, snapshot interactive elements, and interact using the generated element refs.

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 testing without sending the full DOM to the agent?

Browser automation can be run using snapshot-based element references, which capture interactive elements and generate refs for interactions without sending the full DOM. This snapshot-driven workflow enables deterministic page navigation and UI validation.

What's the best way to extract web data and take screenshots for QA testing?

Web data extraction and screenshot capture are handled through automated browser interactions. By navigating pages and snapshotting interactive elements, the agent collects information and validates UI flows for testing, QA, and demos.

How do I fill out web forms automatically using a bash-based agent?

Form filling is automated using Bash-based tooling and agent-browser commands. After opening a page and snapshotting interactive elements, the generated element refs are used to drive form inputs and interactions deterministically.

Does browser automation work with snapshot element references for debugging UI flows?

Browser automation supports debugging and documentation through reference-led workflows. Snapshot-based element references drive interactions and validate UI flows, enabling precise debugging of web pages without manual inspection.

Can I use agent-browser commands to navigate pages and collect information automatically?

Agent-browser commands automate navigating pages and collecting information by opening URLs, snapshotting interactive elements, and interacting using the generated refs. This supports data extraction, testing, and automated UI flow validation.