skills.rest
danfrdn
danfrdnCOMMUNITY
Explore all skills by this author →
GitHub

agent-browser

Automate browser interactions via bash commands using ref-based element selection.

Updated Feb 17, 2026
Search Tags:#screenshots#cli#browser automation#web scraping#form filling#web interaction
One-click install
npx skills add https://github.com/danfrdn/antigravity-config --skill agent-browser-danfrdn
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: agent-browser
Source: https://github.com/danfrdn/antigravity-config/tree/main/skills/agent-browser
Command: npx skills add https://github.com/danfrdn/antigravity-config --skill agent-browser-danfrdn

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill automates interactions with web pages, making it easy to fill forms, take screenshots, scrape data, and perform other browser-based tasks programmatically.

Core Features & Use Cases

  • Web Automation: Navigate websites, click buttons, fill forms, and extract information.
  • Screenshotting & PDF Generation: Capture visual representations of web pages.
  • AI-Friendly Interaction: Uses ref-based element selection from accessibility snapshots, ideal for LLM integration.
  • Use Case: Automatically fill out a login form on a website, navigate to a dashboard, and take a screenshot of the results.

Quick Start

Use the agent-browser skill to open the website "https://example.com" and take a screenshot.

Frequently Asked Questions about agent-browser

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

FAQPage Schema
How do I automate web scraping and form filling from the command line?▼

You can automate web scraping and form filling from the command line by using the agent-browser CLI to execute bash commands that programmatically navigate web pages, interact with elements, and extract data.

Can AI agents use CLI browser automation to take screenshots programmatically?▼

Yes, AI agents can use CLI browser automation to take screenshots programmatically. The agent-browser Skill captures visual representations of web pages and generates PDFs via bash commands.

How does an AI agent interact with web pages for browser automation without visual selectors?▼

For browser automation, an AI agent interacts with web pages by using ref-based element selection derived from accessibility snapshots, allowing precise programmatic control over page elements without visual selectors.

Do I need any external dependencies to run browser automation tasks with bash commands?▼

No external dependencies are required to run browser automation tasks with bash commands. The Skill operates independently to execute form filling, data scraping, and screenshotting directly.

What is the best way to programmatically fill out a login form and capture a dashboard screenshot?▼

The best way to programmatically fill out a login form and capture a dashboard screenshot is using the agent-browser CLI, which is designed to navigate websites, complete form submissions, and take screenshots sequentially.

Are there limitations when using ref-based element selection for web interaction?▼

When using ref-based element selection for web interaction, limitations may arise if the targeted web page lacks proper accessibility snapshots, which are required to accurately map and select elements for programmatic browser manipulation.

Related Skills

Explore More Software Engineering Skills→
sadie100sadie100

scrape

Retrieve text and structured data from web pages via browser automation.

Community
Intermediate
1479089714790897
7

kwp-bio-research-instrument-data-to-allotrope

Convert laboratory instrument output files into Allotrope Simple Model JSON or CSV.

Community
Advanced
anoopg87anoopg87
1

create-agent-graph

Build multi-agent workflow graphs with SimpleAgentGraphBuilder and ConditionalAgentGraphBuilder.

Community
Advanced
MaxStrootmannMaxStrootmann

agent-browser

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

Community
Advanced