What makes this different from standard Playwright scripts?

This skill uses a persistent state and a unique referencing system ('refs') that allows Claude to interact with a page across multiple turns without losing context or session data.

Does the browser run in headless mode?

No, by default, the browser opens visually so you can monitor Claude's interactions in real-time, which is essential for debugging and verification.

Why do I need to run snapshot.js first?

The snapshot script analyzes the current page and assigns temporary IDs (refs) to elements, which Claude then uses to accurately click, type, or hover.

Can this skill handle authenticated sessions?

Yes, because the browser state is persistent between commands, you can navigate to a login page, perform the login interaction, and continue your workflow as an authenticated user.

Playwright Browser Automation

Name: Playwright Browser Automation
Author: VitorAndTxr

byVitorAndTxr

0•

Browser Automation

Automates web interactions, testing, and data collection using persistent browser sessions and element referencing.

The Playwright Skill empowers Claude to perform sophisticated web automation tasks by maintaining a persistent browser state across multiple commands. It features a unique snapshot-based referencing system that allows for precise interaction with DOM elements, facilitating complex workflows like automated form filling, visual regression testing, and dynamic web scraping. Whether you need to generate high-quality screenshots, export pages as PDFs, or create end-to-end test scripts, this skill provides a robust framework for interacting with the modern web directly from your AI development environment.

Key Features

01High-fidelity visual output including screenshots and PDF generation

02Built-in test generation capabilities for rapid E2E development

03Snapshot-based element referencing for reliable UI interaction

04Comprehensive suite of 16 scripts for navigation, input, and capture

05Persistent browser state preservation between command executions

060 GitHub stars

Use Cases

01Automating complex multi-step end-to-end testing workflows

02Scraping data from dynamic, JavaScript-heavy web applications

03Capturing visual documentation and UI audit snapshots automatically

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add vitorandtxr/bioteca.prism.interoperableresearchinterfacesystem playwright

For use in Claude.ai and ChatGPT

Download Skill

Key Features

01High-fidelity visual output including screenshots and PDF generation

02Built-in test generation capabilities for rapid E2E development

03Snapshot-based element referencing for reliable UI interaction

04Comprehensive suite of 16 scripts for navigation, input, and capture

05Persistent browser state preservation between command executions

060 GitHub stars

Use Cases

01Automating complex multi-step end-to-end testing workflows

02Scraping data from dynamic, JavaScript-heavy web applications

03Capturing visual documentation and UI audit snapshots automatically

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add vitorandtxr/bioteca.prism.interoperableresearchinterfacesystem playwright

For use in Claude.ai and ChatGPT

Download Skill