skills.rest
avoidthekitchen
avoidthekitchenCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

agent-browser

Automate Chrome/Chromium browser interactions via the DevTools Protocol.

1|Updated Apr 13, 2026
Search Tags:#cdp#session-management#e2e-testing#web-scraping#browser-automation#form-filling#video-recording
One-click install
npx skills add https://github.com/avoidthekitchen/avilingo-v2 --skill agent-browser-avoidthekitchen
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: agent-browser
Source: https://github.com/avoidthekitchen/avilingo-v2/tree/main/.agents/skills/agent-browser
Command: npx skills add https://github.com/avoidthekitchen/avilingo-v2 --skill agent-browser-avoidthekitchen

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill provides a reliable, CLI-driven way for AI agents and engineers to interact programmatically with websites and desktop web apps, removing the manual toil of navigation, form entry, scraping, and exploratory QA while preserving session state and authentication.

Core Features & Use Cases

  • Browser automation CLI: Control Chrome/Chromium via the DevTools Protocol for navigation, clicking, typing, file upload, and complex interaction sequences.
  • Snapshot-driven interactions: Accessibility-tree snapshots produce compact element refs for low-context, robust actions that avoid brittle CSS selectors.
  • Session & authentication management: Save and load session state, use persistent profiles or an auth vault, and handle OAuth, 2FA, and token refresh workflows.
  • Specialized workflows: Templates and references for authenticated sessions, form automation, content capture, profiling, proxy support, video recording, Electron and Slack automation, and running in cloud sandbox providers.
  • Use Case: Automate a login, run a purchase flow or regression smoke test, capture screenshots and a video recording, and save authenticated state for repeated CI or manual runs.

Quick Start

Use the agent-browser CLI to install and load the skill, open the target URL, run an interactive snapshot to discover element refs, and then perform the desired automated interactions using those refs.

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 without relying on brittle CSS selectors?▼

You can automate browser interactions using accessibility-tree snapshots that produce compact element refs for low-context actions, avoiding brittle CSS selectors. This method allows reliable clicking, typing, and navigation directly via the DevTools Protocol in Chrome or Chromium.

What is the best way to handle authentication and session state for end-to-end testing?▼

Handling authentication for end-to-end testing involves using session and state management to save and load session state. You can use persistent profiles or an auth vault to manage OAuth, 2FA, and token refresh workflows for repeated automated runs.

Can I use the DevTools Protocol for web scraping and video recording?▼

Yes, the DevTools Protocol supports web scraping and video recording by controlling Chrome or Chromium directly. You can capture content, take screenshots, record video sessions, and execute complex scraping sequences programmatically.

Does browser automation work with Electron and Slack desktop applications?▼

Yes, browser automation works with Electron and Slack desktop applications via the DevTools Protocol. It provides specialized workflows and templates for desktop automation, allowing programmatic navigation, form filling, and UI testing on these platforms.

How do I run automated UI testing in a cloud sandbox environment?▼

To run automated UI testing in a cloud sandbox, use a CLI-first workflow that supports cloud sandbox provider execution. This enables you to run regression smoke tests, manage profiles, and capture screenshots within isolated, scalable cloud environments.

Why use accessibility-tree snapshots instead of traditional DOM queries for form filling?▼

Accessibility-tree snapshots are used over traditional DOM queries for form filling because they produce compact element refs for robust, low-context actions. This method avoids brittle CSS selectors, ensuring reliable automation during complex interaction sequences.

Related Skills

Explore More Software Engineering Skills→
babuallybabually
1

agent-browser

Control Chrome or Chromium via CLI using accessibility-tree snapshots and element references.

Community
Advanced
PostHogPostHog
58

diagnosing-missing-recordings

Diagnose missing session recordings using PostHog diagnostic signals and decision trees.

Official
Advanced
8b-is8b-is

agent-browser

Automate browser navigation, element interaction, and data extraction via CLI.

Official
Advanced
kompassdevkompassdev
9

agent-browser

Automate web browser navigation, interaction, data extraction, and testing workflows.

Official
Intermediate