agent-browser

Automate browser navigation, form filling, and screenshot capture via DOM refs.

1|2|Updated Apr 8, 2026
One-click install
npx skills add https://github.com/jboothomas/nauggieclaw --skill agent-browser-jboothomas
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-browser
Source: https://github.com/jboothomas/nauggieclaw/tree/main/container/skills/agent-browser
Command: npx skills add https://github.com/jboothomas/nauggieclaw --skill agent-browser-jboothomas

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Oversees complex web navigation tasks by letting the agent operate a browser so users no longer need to manually research, interact with forms, or collect data from web applications.

Core Features & Use Cases

  • Guided navigation: open URLs, go back, forward, reload, and close the browser to explore any sequence of pages safely within the container.
  • Interactive snapshots and DOM refs: capture accessibility trees or interactive elements with stable references to click, type, select, and hover without guessing selectors.
  • Data extraction and verification: read text, HTML, attributes, page titles, and URLs; take screenshots or PDFs for evidence; save and reuse authentication state for repeated sessions.
  • Use Case: Walk through a login flow by opening the login page, filling saved credentials via refs, clicking submit, waiting for the dashboard URL, and then capturing a screenshot.

Quick Start

Start with agent-browser open followed by the URL you want to inspect so the skill can load the page and provide references for immediate interaction.

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 navigation and form filling without guessing CSS selectors?

Browser automation uses interactive snapshots and stable DOM references to click, type, and select elements. This lets you automate form filling and web navigation without guessing selectors or manually inspecting HTML.

Can I capture screenshots and extract data from complex web apps automatically?

Yes, you can capture screenshots or PDFs for evidence and extract data by reading text, HTML, attributes, page titles, and URLs. This automates data extraction from complex web apps during interactive testing.

How do I handle login flows and authentication state when automating browser tasks?

You automate login flows by opening the page, filling credentials via DOM refs, and clicking submit. The browser saves and reuses authentication state so you can repeat sessions without logging in manually.

What is the best way to verify UI state during automated web navigation?

The best way to verify UI state is capturing interactive snapshots and screenshots during web navigation. You can read page titles, URLs, and element attributes to confirm the application rendered correctly after interactions.

Do I need command-driven browser tooling to run JavaScript snippets and manage cookies?

Yes, this approach requires command-driven browser tooling that can open URLs, snapshot DOMs, interact via refs, manage cookies, and run JavaScript snippets within a safe container to operate complex web tasks.