skills.rest
haidonglethqb
haidonglethqbCOMMUNITY
Explore all skills by this author →
GitHub

agent-browser

Automate browser tasks for AI agents using a snapshot and refs CLI.

Updated Feb 27, 2026
Search Tags:#automation#cli#ai#cloud#browser#snapshots
One-click install
npx skills add https://github.com/haidonglethqb/CloudSchool --skill agent-browser-haidonglethqb
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: agent-browser
Source: https://github.com/haidonglethqb/CloudSchool/tree/main/.claude/skills/agent-browser
Command: npx skills add https://github.com/haidonglethqb/CloudSchool --skill agent-browser-haidonglethqb

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

AI agents often struggle to automate web tasks efficiently because traditional browser automation requires large context and brittle scripts. agent-browser offers a CLI with a snapshot-and-refs workflow to minimize context while maintaining reliable interactions.

Core Features & Use Cases

  • Snapshot + refs workflow enables context-efficient browser automation for long autonomous sessions.
  • Built-in video recording, self-verifying workflows, and optional cloud testing with Browserbase support.
  • Cloud-friendly and extensible: supports local browsers and cloud providers to scale tests.

Quick Start

Install agent-browser globally and begin by opening a URL, taking a snapshot, and performing an interaction sequence.

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 for AI agents without consuming too much context?▼

Browser automation for AI agents can minimize context by using a snapshot plus refs workflow. This model takes a snapshot of the page state and uses reference tags for interactions, reducing the context needed for long autonomous sessions.

What is the snapshot and refs model for browser automation?▼

The snapshot and refs model captures the current browser state as a snapshot and assigns reference tags to interactive elements. AI agents use these refs to navigate and interact, maintaining reliable workflows without processing raw HTML.

Can I run self-verifying browser automation workflows in the cloud?▼

Yes, self-verifying browser automation workflows can run in the cloud using Browserbase support. This allows you to scale tests and run long autonomous sessions on cloud browsers while recording video of the interactions.

How do I start using a CLI for AI browser automation?▼

To start AI browser automation via CLI, install the agent-browser tool globally, open a target URL, take a snapshot of the page, and execute your interaction sequence using the generated element references.

Does AI browser automation support network control and state persistence?▼

AI browser automation supports network control and state persistence to maintain long autonomous sessions. These features allow agents to manage network requests and save session states across multiple interactions.

Why do traditional browser automation scripts fail for long AI sessions?▼

Traditional browser automation scripts fail for long AI sessions because they require large context windows and rely on brittle selectors. A snapshot and refs workflow minimizes context and provides stable references for reliable interactions.

Related Skills

Explore More Software Engineering Skills→
congthang12312congthang12312

agent-browser

Automate browser tasks via CLI with context-aware snapshots.

Community
Advanced
benjaminshoemakerbenjaminshoemaker
39

analyze-sessions

Analyze Claude Code session logs to identify automation opportunities.

Community
Basic
8b-is8b-is

agent-browser

Automate browser navigation, element interaction, and data extraction via CLI.

Official
Advanced
MICHAEL-BODOMICHAEL-BODO
2

browser-automation

Automate browser interactions for web testing and scraping with Playwright and Puppeteer.

Community
Intermediate