agent-browser

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

22|3|Updated Feb 22, 2026
One-click install
npx skills add https://github.com/DFilipeS/praxis --skill agent-browser-dfilipes
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-browser
Source: https://github.com/DFilipeS/praxis/tree/main/praxis/skills/agent-browser
Command: npx skills add https://github.com/DFilipeS/praxis --skill agent-browser-dfilipes

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Automate repetitive browser tasks, enabling AI-assisted agents to navigate, fill forms, extract data, and verify web apps without manual scripting.

Core Features & Use Cases

  • Comprehensive command set for navigation, interaction, data extraction, testing, and documentation of automation flows.
  • Supports multi-session isolation, state persistence, and ref-based interactions for robust parallel automation and authenticated workflows.
  • Real-world use cases include form automation, web testing, data scraping, and automated QA documentation.

Quick Start

Open a target page, snapshot to identify element refs, and begin automating interactions.

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 for AI-driven workflows without manual coding?

Automating browser tasks for AI workflows is achieved by using commands to navigate pages, fill forms, extract data, and test web apps without manual scripting. You open a target page, snapshot to identify element refs, and begin automating interactions.

How does snapshot and ref-based interaction work for web testing?

Snapshot and ref-based interaction works by capturing a snapshot of a target page to identify specific element references. You then use these refs to directly automate interactions like clicking buttons or filling forms across multiple sessions and devices.

Can I run parallel automation and maintain authenticated sessions across multiple devices?

Yes, you can run parallel automation and maintain authenticated sessions across multiple devices. The automation supports multi-session isolation and state persistence, ensuring robust parallel workflows and authenticated states are maintained.

What's the best way to extract data and scrape web pages using AI agents?

The best way to extract data and scrape web pages using AI agents is to apply the command set for data extraction and navigation. You navigate to target pages, snapshot the DOM to identify element refs, and automate the data extraction process.

Does this browser automation approach support automated QA documentation and web app testing?

Yes, this browser automation approach supports automated QA documentation and web app testing. It provides a comprehensive command set for testing web apps and documenting automation flows, satisfying requirements for verifying web applications across sessions.

What are the limitations of using AI agents for web automation and form filling?

Limitations of using AI agents for web automation include the need to snapshot pages to identify element refs before interacting, and the requirement to structure flows using YAML frontmatter with a name and description for on-demand script execution.