claude-bowser

Automate observable Chrome browser actions using MCP tools.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/declanshanaghy/fenrir-ledger --skill claude-bowser-declanshanaghy
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: claude-bowser
Source: https://github.com/declanshanaghy/fenrir-ledger/tree/main/.claude/skills/claude-bowser
Command: npx skills add https://github.com/declanshanaghy/fenrir-ledger --skill claude-bowser-declanshanaghy

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Automate observable browsing tasks by using Chrome MCP tools in your real Chrome instance, enabling tests and data collection without headless modes or external services.

Core Features & Use Cases

  • Observable browser automation: operate on your active Chrome profile with real cookies and extensions.
  • Deterministic task execution: resize window, navigate, interact, and capture results in a predictable flow.
  • Use Case: perform browser-based QA checks or collect screenshots across multiple sites for reporting.

Quick Start

Start Claude with the Chrome flag and issue your browser task; the workflow will verify tool availability and run in the current Chrome session.

Frequently Asked Questions about claude-bowser

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

FAQPage Schema
How do I automate browser tasks in my real Chrome session?

You can automate observable browser actions in Chrome by using MCP tools to manipulate and inspect web pages. This skill operates on your active Chrome profile to perform tasks like website QA and content verification without headless modes.

Can I use my real Chrome cookies and extensions during UI testing?

Yes, observable browser automation operates directly on your active Chrome profile. This allows you to run automated browsing workflows and UI testing with your real cookies and extensions intact, ensuring accurate content verification.

How do I capture screenshots across multiple sites for QA reporting?

You can collect screenshots across multiple sites by issuing a browsing task in your current Chrome session. The workflow verifies MCP tool availability, navigates deterministically, interacts with web pages, and captures results for reporting.

Do I need Chrome MCP tools available to automate web page inspection?

Yes, Chrome MCP tools are required to automate web page inspection and manipulation. The workflow verifies that these MCP tools are installed and available before executing any deterministic, retryable browsing tasks.

What is the best way to perform deterministic website QA checks?

The best way to perform deterministic website QA is by automating observable browser actions in Chrome via MCP tools. This approach resizes windows, navigates, interacts, and captures results in a predictable, retryable flow without external services.

Why use observable Chrome automation instead of headless browser modes?

Observable Chrome automation is used instead of headless modes to enable tests and data collection on your active profile. This approach utilizes your real cookies and extensions, providing a more accurate environment for website QA and content verification.