agent-browser

Automate browser navigation and interaction via CLI actions like open, snapshot, fill, and click.

Updated Mar 1, 2026
One-click install
npx skills add https://github.com/Zoppy-crm/.github --skill agent-browser-zoppy-crm
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-browser
Source: https://github.com/Zoppy-crm/.github/tree/main/skills/frontend/agent-browser
Command: npx skills add https://github.com/Zoppy-crm/.github --skill agent-browser-zoppy-crm

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Browser automation for AI agents to programmatically interact with websites, fill forms, navigate pages, capture screenshots, and extract data, reducing manual browser tasks.

Core Features & Use Cases

  • Navigate and interact with web pages using a CLI to perform repetitive tasks without manual clicks.
  • Snapshot and generate element refs for deterministic interactions, enabling reliable automation across sessions.
  • Use cases include form submission, data extraction, testing, authentication flows, and content capture.

Quick Start

Open a website with the agent, snapshot to discover refs, then fill or click using refs to achieve a task.

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 form filling without manual clicks?

You can automate web form filling by using a CLI to open the target page, snapshot to discover element refs, and issue fill actions using those refs for deterministic, repeatable form submission.

What is snapshot ref generation in browser automation?

Snapshot ref generation captures web page elements and generates deterministic references, allowing AI agents to reliably target and interact with specific elements like forms or buttons across different sessions.

Can I use a CLI to extract data and capture screenshots from web pages?

Yes, CLI workflows support data extraction and screenshot capture by navigating web pages programmatically, returning context-ready instructions and visual outputs for downstream processing.

Does browser automation work for testing authentication flows?

Browser automation handles authentication flows by programmatically navigating login pages, filling credentials, and interacting with elements using deterministic refs to test secure access scenarios reliably.

What are the limitations of deterministic refs in web testing?

Deterministic refs rely on snapshot element discovery, meaning dynamic content or rapidly changing DOM structures may require fresh snapshots to maintain reliable interaction during web testing automation.