agent-browser

Consolidate browser automation into the /carmack agent with CLI commands.

5|Updated Mar 27, 2026
One-click install
npx skills add https://github.com/barkleesanders/claude-code-starter --skill agent-browser-barkleesanders
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-browser
Source: https://github.com/barkleesanders/claude-code-starter/tree/main/skills/agent-browser
Command: npx skills add https://github.com/barkleesanders/claude-code-starter --skill agent-browser-barkleesanders

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill removes the need for a separate browser automation tool by merging all its functionality into the /carmack agent, giving users access to all browser interaction capabilities through a single unified engineering assistant.

Core Features & Use Cases

  • Full Browser CLI Command Set: Includes commands for opening pages, taking snapshots, clicking elements, filling forms, typing text, scrolling, and waiting for elements to load.
  • Advanced Browser Controls: Supports semantic element locators, session persistence, JSON output, video recording, debug (--headed) mode, cookie and storage management, and network interception.
  • Common Use Cases: Automate form submission, handle authenticated workflows with saved session state, and run parallel browser sessions for testing or data collection.

Quick Start

Ask the /carmack agent to complete your browser automation task, such as logging into a website and extracting data from a dashboard.

Frequently Asked Questions about agent-browser

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate form submission and authenticated web scraping in a CLI environment?

Browser automation for form submission and authenticated web scraping is handled through a unified CLI agent, providing commands to open pages, fill forms, and manage persistent sessions. It eliminates the need for a separate dedicated browser tool.

Can I run parallel browser sessions for testing or data collection?

Yes, parallel browser sessions are supported for testing and data collection tasks. The unified agent manages multiple sessions simultaneously, handling persistent state and authenticated workflows across parallel browser instances.

Does browser automation work with network interception and semantic element locators?

Network interception and semantic element location are core features of this browser automation approach. It intercepts network requests and locates web elements semantically, outputting results in JSON format for debugging and analysis.

How do I manage cookies and storage state for persistent browser sessions?

Cookie and storage management commands allow you to save and reuse authenticated session state. This enables persistent browser sessions that maintain login credentials across multiple automation runs without re-authenticating.

What is debug mode browser interaction and when do I need it?

Debug mode browser interaction, or headed mode, launches a visible browser window for troubleshooting automation scripts. It is needed when diagnosing form submission failures or verifying semantic element locators during development.

Do I need a separate browser automation tool if I already use an engineering agent?

No, a separate browser automation tool is unnecessary if your engineering agent consolidates this functionality. The /carmack agent merges browser CLI commands, session persistence, and network interception into a single interface.