skills.rest
roach88
roach88COMMUNITY
Explore all skills by this author →
GitHub

agent-browser

Automate web interactions via the agent-browser CLI to navigate, fill forms, and gather data.

3|Updated Feb 5, 2026
Search Tags:#cli#web-scraping#browser-automation#agent-browser#accessibility-snapshot#ref-based-selection
One-click install
npx skills add https://github.com/roach88/compound-engineering --skill agent-browser-roach88
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: agent-browser
Source: https://github.com/roach88/compound-engineering/tree/main/skills/agent-browser
Command: npx skills add https://github.com/roach88/compound-engineering --skill agent-browser-roach88

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Browser automation often requires manual, repetitive steps across web interfaces. agent-browser provides a lightweight CLI path to control a headless browser, enabling agents to interact with pages, fill forms, take screenshots, and scrape data efficiently.

Core Features & Use Cases

  • Navigate pages and open URLs via simple commands.
  • Snapshot interactive elements with refs for deterministic automation.
  • Quick data extraction and form interactions for automation workflows.

Quick Start

Open a URL and create an interactive element snapshot to identify refs for automation.

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 scraping using a CLI browser?▼

The agent-browser CLI automates web interactions by navigating pages and discovering interactive elements through accessibility snapshots. It uses ref-based selection to target page elements for navigation and data extraction without traditional CSS selectors.

How does ref-based selection work for browser automation?▼

Ref-based selection uses accessibility snapshots to identify interactive elements on web pages. This allows scripts to target page components deterministically by their refs instead of relying on fragile DOM selectors for form interactions.

Can I integrate CLI web scraping outputs into AI workflows?▼

Yes, browser automation tasks output results in JSON format. This structured output makes extracted web data immediately ready for integration into downstream AI workflows and multi-session scripting environments.

Do I need a specific browser dependency to run agent-browser?▼

You need the agent-browser CLI installed to run this tool. It operates in a Bash environment and controls a headless browser without requiring additional external dependencies or complex component installations.

What are the limitations of using a CLI for browser automation?▼

CLI browser automation is designed for scripting in Bash environments and may not support visual debugging. It focuses on headless navigation and JSON outputs, making it less suited for workflows requiring real-time visual interaction.

How do I validate web page interactions during headless automation?▼

You validate interactions using snapshot-based validation to verify page state. The CLI takes accessibility snapshots of interactive elements, allowing you to confirm navigation and form filling success before extracting data.

Related Skills

Explore More Software Engineering Skills→
marktantongcomarktantongco

playwright-cli

Automate headless browser navigation, clicks, and form submissions for testing and scraping.

Community
Intermediate
archibatearchibate
6

slack

Automate Slack web workspace interactions to read messages and extract channel metadata.

Community
Intermediate
skuridinskuridin

playwright-cli

Automate browser navigation, form interaction, screenshots, and network mocking with Playwright.

Community
Intermediate
szsip239szsip239
103

Agent Browser

Automate headless browser interactions via CLI commands like open, click, and snapshot.

Community
Intermediate