browser-puppet

Automate Playwright browser tasks and capture evidence via an MCP server.

Updated Mar 22, 2026
One-click install
npx skills add https://github.com/bja2142/vibes --skill browser-puppet
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: browser-puppet
Source: https://github.com/bja2142/vibes/tree/main/browser-mcp
Command: npx skills add https://github.com/bja2142/vibes --skill browser-puppet

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Browser automation tasks in real-world web apps are hard to reproduce, test, and observe consistently without an integrated runtime. This skill provides an MCP-powered solution to drive Playwright-backed browsers, collect evidence, and maintain per-context state for reliable automation.

Core Features & Use Cases

  • Stateful browser contexts and page tracking for end-to-end testing
  • Inline element resolution and post-action observation for efficient workflows
  • Evidence capture (screenshots, HAR, traces, video) and data extraction across desktop and mobile scenarios
  • Use Case: validate a multi-step user registration flow, including network/console checks and dynamic UI updates.

Quick Start

Open a browser context, navigate to a URL, and start automating interactions.

Frequently Asked Questions about browser-puppet

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

FAQPage Schema
How do I automate end-to-end UI testing with Playwright and capture evidence?

Playwright automation can capture evidence like screenshots, HAR, traces, and video by driving real browsers via an MCP server. It maintains stateful browser contexts and page tracking for reliable end-to-end UI testing workflows.

What is the best way to handle complex authentication flows during browser automation?

Browser automation can handle complex auth flows using lightweight credential management and per-context state maintenance. This allows automated real-browser tasks to securely navigate authentication sequences across desktop and mobile scenarios.

Can I use MCP server architecture for data extraction across desktop and mobile contexts?

Yes, an MCP server can orchestrate real browsers for data extraction across desktop and mobile contexts. It uses context and page discovery alongside inline element resolution to efficiently extract data during automated interactions.

Does browser automation allow guarded mutations with post-action observation?

Browser automation supports guarded mutation tools with optional post-action observation. This mechanism enables efficient workflows by allowing controlled state changes while observing dynamic UI updates immediately after the action.

How do I validate a multi-step user registration flow with network and console checks?

Validating a multi-step registration flow involves opening a browser context, navigating to the URL, and automating interactions. The MCP runtime collects network and console checks while capturing evidence of dynamic UI updates throughout the flow.