Browser4
AI-driven browser automation, web scraping, and data extraction at scale
All Skills in This Repository (21)
Pure Emerald Level Indicatorsscript-runner
Validate Python script execution pipelines for allowed tools, timeouts, and output limits.
Extract text and structured data from PDF documents using pypdf and pdfplumber.
form-filling
Automate web form input and submission with provided data.
weather
Fetch real-time weather and 7-day forecasts for locations via Open-Meteo APIs.
web-scraping
Extract titles, text, and structured fields from web pages using browser automation.
data-validation
Validate data against predefined and custom rules.
mcp-builder
Develop MCP servers for LLM integration with Python and Node/TypeScript SDKs.
task-token-usage
Analyzes Claude Code session traces to report token usage per task.
organize-task-files
Organizes Coworker task files across pipeline stages by scanning and parsing them automatically.
browser4-cli
Automate browser interactions for web testing, form filling, screenshots, and data extraction.
coworker
Automates file-queue task pipelines using PowerShell workers and GitHub Copilot.
test-production
Acceptance-tests the published browser4-cli release by installing, exercising, and uninstalling it.
Frequently Asked Questions
FAQPage SchemaHow to install Browser4?โผ
Run `npx skills add platonai/Browser4 --all -g -y` in your terminal to install all skills in this suite globally.
What can Browser4 automate in a browser?โผ
It navigates pages, clicks and fills forms, takes screenshots, handles dialogs, and extracts structured data using CSS selectors or X-SQL queries. AI agents can also submit plain-English tasks and let the built-in agent plan and execute the browser steps autonomously.
Can Browser4 scrape data without using LLM tokens?โผ
Yes. X-SQL queries and CSS selectors extract data deterministically, and the WebMiner tool clusters downloaded HTML pages into spreadsheets and reports entirely locally with zero LLM tokens.
Does Browser4 work with Claude Code and other AI agents?โผ
Yes. It exposes browser control through MCP over HTTP and follows the standard SKILL.md format, so it works with Claude Code, Codex, OpenClaw, and similar agents.
Can Browser4 handle large-scale crawling?โผ
Yes. Its swarm mode runs parallel browser contexts designed for 100k to 200k page visits per machine per day, with plugins for CAPTCHA solving, SEO audits, and media processing.
Related Repositories in Software Engineering
View All in Software Engineeringโopenclaw
Run a personal AI assistant across your devices and chat apps
superpowers
Gives coding agents a disciplined workflow from idea to merged code
react
AI agent skills for building, testing, and porting React core