claude-tweaks:browse

Routes browser tasks between Playwright CLI and Chrome MCP backends automatically.

2|Updated Feb 20, 2026
One-click install
npx skills add https://github.com/thomasholknielsen/claude-tweaks --skill claude-tweaks-browse
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: claude-tweaks:browse
Source: https://github.com/thomasholknielsen/claude-tweaks/tree/main/skills/browse
Command: npx skills add https://github.com/thomasholknielsen/claude-tweaks --skill claude-tweaks-browse

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Unified browser automation that automatically selects the best available backend (Playwright CLI or Chrome MCP) and routes commands accordingly, simplifying cross-backend tasks.

Core Features & Use Cases

  • Backend auto-detection & routing: picks Playwright CLI or Chrome MCP based on availability.
  • Headless parallel Playwright sessions: enables fast, scalable browser tasks.
  • Observable Chrome workflows: real-browser interactions for QA and testing.
  • Cross-skill usage: leveraged by /claude-tweaks:stories and /claude-tweaks:review for browser tasks.

Quick Start

Provide a URL and task description to start browser automation; the system will select the best backend and execute the task.

Frequently Asked Questions about claude-tweaks:browse

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

FAQPage Schema
How do I automate browser tasks across different automation backends?

Browser automation across different backends is handled by automatically detecting whether Playwright CLI or Chrome MCP is available and routing commands to the best option. This simplifies cross-backend tasks without manual configuration.

Can I run parallel Playwright sessions for web automation?

Yes, you can run parallel Playwright sessions for web automation. The system supports headless parallel Playwright sessions, enabling fast and scalable browser tasks for data scraping or UI testing.

What is the best way to handle UI testing with observable Chrome sessions?

For UI testing with observable Chrome sessions, the system routes commands to Chrome MCP, providing real-browser interactions. This approach is suitable for visual QA and testing across web apps.

Does this browser automation approach require Playwright or Chrome MCP to be pre-installed?

Browser automation relies on backend auto-detection, meaning it selects Playwright CLI or Chrome MCP based on availability. You need at least one of these backends present in your environment for command routing to function.

How does web automation backend detection work for interactive browsing?

Web automation backend detection works by evaluating the availability of Playwright CLI and Chrome MCP in the environment. It then picks the best available backend and maps operations accordingly for interactive browsing tasks.

Can I use browser automation for visual QA across web apps?

Yes, you can use browser automation for visual QA across web apps. It supports observable Chrome workflows for real-browser interactions and headless Playwright sessions, ending with standardized reporting of the results.