skills.rest
Shirataki2
Shirataki2COMMUNITY
Explore all skills by this author →
GitHub

agent-browser

Automate browser navigation, form filling, clicking, and data extraction via CLI.

2|Updated Dec 1, 2025
Search Tags:#automation#testing#cli#session-management#data-extraction#browser-automation#web-interactions
One-click install
npx skills add https://github.com/Shirataki2/discalendar-next-v2 --skill agent-browser-shirataki2
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: agent-browser
Source: https://github.com/Shirataki2/discalendar-next-v2/tree/main/.agents/skills/agent-browser
Command: npx skills add https://github.com/Shirataki2/discalendar-next-v2 --skill agent-browser-shirataki2

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Automating browser-based tasks is tedious and error-prone when done manually, especially for AI agents that must interact with websites, fill forms, navigate pages, and validate results.

Core Features & Use Cases

  • Snapshotting: identify interactive elements with short, context-rich refs to drive AI actions.
  • Interaction primitives: open/navigate, fill, click, type, select, and wait for network or URL changes.
  • Session management: isolate multiple browser contexts and persist state for reuse across tasks.
  • Data extraction and validation: extract text or metadata and verify outcomes across pages.
  • Use Case: automate login flows, form submissions, and data collection from multiple sites with reproducible steps.

Quick Start

Open a target website with agent-browser open https://example.com and then run a snapshot to discover element refs before performing automated actions.

Frequently Asked Questions about agent-browser

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

FAQPage Schema
How do I automate browser interactions for AI agents?▼

You can automate browser interactions for AI agents using a CLI that handles navigation, element identification with refs, form filling, clicking, screenshots, and data extraction across multiple sessions.

What is the best way to manage multiple browser sessions for web automation?▼

The best way to manage multiple browser sessions is to use a CLI that isolates browser contexts and persists state, allowing you to reuse sessions across different automation tasks and websites.

How do I extract data and validate outcomes from web pages automatically?▼

Extract data and validate outcomes by taking snapshots to identify elements, performing actions like fill and click, then extracting text or metadata to verify results across pages.

How do I identify interactive elements on a web page for automation scripts?▼

Identify interactive elements by running a snapshot command, which generates short, context-rich refs for each element to drive subsequent AI actions like filling forms or clicking buttons.

Can I automate login flows and form submissions with reproducible steps?▼

Yes, you can automate login flows and form submissions with reproducible steps by using CLI commands to open URLs, snapshot elements, fill credentials, click submit, and validate the outcome.

What are the limitations of using a CLI for browser automation?▼

Limitations of CLI browser automation include the need to explicitly snapshot pages to discover element refs before acting, and managing state persistence manually when orchestrating multiple isolated sessions.

Related Skills

Explore More Software Engineering Skills→
fbrierfbrier

agent-browser

Automate browser interactions for web testing, form filling, and data extraction.

Community
Advanced
aws-samplesaws-samples
90

file-processing

Process and analyze CSV, JSON, and text data files with automated cleaning and transformation.

Official
Advanced
MinhDuyDEVMinhDuyDEV

playwright

Automate browser testing and UI validation with Playwright workflows.

Community
Advanced
tharunramagiritharunramagiri

agent-browser

Automate web navigation, element interaction, and data extraction via a headless browser.

Community
Advanced