agent-browser

Automate browser navigation, form filling, and element interaction with snapshot refs.

232|14|Updated Jan 26, 2026
One-click install
npx skills add https://github.com/dmmulroy/overseer --skill agent-browser-dmmulroy
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-browser
Source: https://github.com/dmmulroy/overseer/tree/main/.agents/skills/agent-browser
Command: npx skills add https://github.com/dmmulroy/overseer --skill agent-browser-dmmulroy

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Automates repetitive browser interactions and data extraction, reducing manual testing and scraping effort across websites.

Core Features & Use Cases

  • Automated Navigation & Interaction: Open URLs, click elements, fill forms, take screenshots, and capture page structure with compact refs to minimize AI token usage.
  • Snapshot Ref System: Generate and reuse short element references to drive actions without full DOM payloads, improving efficiency for AI agents.
  • State Management & Debugging: Support parallel sessions, state persistence, and logging for reproducible tests and demos.

Quick Start

Open a URL, snapshot interactive elements, and perform actions using element refs to automate a test flow.

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 testing without sending large DOM payloads to AI agents?

Automate web testing with a snapshot ref system that generates compact element references, allowing AI agents to drive browser interactions using minimal tokens instead of full DOM payloads.

What is the best way to automate browser scraping and form filling for repetitive web tasks?

Automate browser scraping and form filling by opening URLs, snapshotting interactive elements, and performing actions through short element refs to streamline navigation, form automation, and content gathering.

Can I run parallel browser automation sessions and capture state for reproducible tests?

Parallel browser automation sessions are supported alongside state persistence and logging, enabling you to capture state and create reproducible tests and demos across websites and web apps.

Does browser automation with CDP require external dependencies or packages?

Browser automation with CDP requires no external dependencies, operating directly through a YAML frontmatter configuration in SKILL.md with optional scripts, references, and assets directories.

Why use compact element refs for web scraping instead of standard DOM parsing?

Compact element refs minimize AI token usage during web scraping by generating short references that replace full DOM payloads, improving efficiency when AI agents drive navigation and data extraction.