skills.rest
miklosme
miklosmeCOMMUNITY
Explore all skills by this author →
𝕏@miklosmeGitHub

agent-browser

Automate Chromium-based browser interactions using snapshot-derived element refs.

Updated Apr 10, 2026
Search Tags:#ai-agents#web-scraping#browser-automation#form-filling#multisession#ref-based-interactions
One-click install
npx skills add https://github.com/miklosme/nextjs-ai-bootstrap --skill agent-browser-miklosme
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: agent-browser
Source: https://github.com/miklosme/nextjs-ai-bootstrap/tree/main/.agents/skills/agent-browser
Command: npx skills add https://github.com/miklosme/nextjs-ai-bootstrap --skill agent-browser-miklosme

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Browser automation for AI agents that need to interact with websites, including navigation, form filling, clicking, screenshots, data extraction, and web app testing.

Core Features & Use Cases

  • Compact, ref-based interactions via snapshot-derived element refs to minimize prompts.
  • Multi-session isolation and state management for concurrent automations.
  • Authentication and data extraction workflows with robust wait strategies and error handling.

Quick Start

Open a target website, snapshot the page to identify refs, fill credentials, submit, and verify successful navigation.

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 tasks like form filling and web scraping with AI agents?▼

Automate browser tasks by using a ref-based model to control a Chromium-based browser, allowing AI agents to navigate pages, fill forms, and scrape data via snapshot-derived element refs.

How does ref-based interaction work for web automation?▼

Ref-based interaction works by taking a snapshot of the page to identify deterministic element refs (@e), which the AI agent then targets with CLI commands to minimize prompt size and execute precise actions.

Can I run concurrent browser automations across multiple sessions?▼

Yes, you can run concurrent browser automations using multi-session isolation and state management, allowing simultaneous execution of independent login flows and web scraping tasks.

Does this browser automation approach support proxy and authentication workflows?▼

Yes, the browser automation approach supports optional state, proxy, and authentication configurations, enabling robust execution of complex login flows and authenticated web app testing.

What is the best way to handle waiting and errors during automated web interactions?▼

The best way to handle waiting and errors is utilizing built-in robust wait strategies and error handling mechanisms, ensuring page elements are fully loaded before executing snapshot-based interactions.

Are there limitations when using snapshot-derived element references for web scraping?▼

Snapshot-derived element references require elements to be present in the DOM snapshot, meaning highly dynamic content may need explicit waiting strategies before the snapshot is taken to ensure accurate scraping.

Related Skills

Explore More Software Engineering Skills→
ruska-airuska-ai
17

agent-browser

Automate browser navigation, form filling, screenshots, and data extraction via agent-browser CLI.

Official
Advanced
Magnatrix-LabMagnatrix-Lab
1

competitive-ads-extractor

Extract competitor ad insights via web scraping and structured data processing.

Community
Advanced
farzanmrzfarzanmrz
6

agent-browser

Automate browser UI tasks via CLI for testing and data extraction.

Community
Advanced
platonaiplatonai
1.1k

browser4-cli

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

Community
Advanced