agent-browser

Automate browser interactions for websites and desktop apps via CDP.

8|Updated Jan 19, 2026
One-click install
npx skills add https://github.com/OiAnthony/.agents --skill agent-browser-oianthony
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-browser
Source: https://github.com/OiAnthony/.agents/tree/main/skills/agent-browser
Command: npx skills add https://github.com/OiAnthony/.agents --skill agent-browser-oianthony

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It removes the friction of manual browser work by letting an AI agent navigate sites, interact with interfaces, extract data, and verify behavior through a fast command-line workflow.

Core Features & Use Cases

  • Website Interaction: Open pages, click elements, fill forms, log in, and capture screenshots without brittle manual steps.
  • Testing and QA: Run exploratory testing, bug hunts, and app checks across web apps and desktop apps with browser-like interfaces.
  • Specialized Automation: Handle Slack tasks, Electron apps, Vercel Sandbox sessions, and AWS Bedrock AgentCore cloud browsers when the task goes beyond standard web browsing.

Quick Start

Load the core workflow with agent-browser skills get core before running any browser automation task.

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 interaction for web testing and data extraction?

You can automate browser interaction by loading the core workflow and using AI to navigate sites, click elements, fill forms, and extract data. It relies on accessibility-tree snapshots and CDP-based browser access for reliable web testing.

Does browser automation work with Electron apps and Slack tasks?

Yes, browser automation supports specialized workflows for Electron apps and Slack tasks. It routes specialized skills to handle interactions beyond standard web browsing, including Vercel Sandbox and AgentCore cloud browsers.

What do I need to set up before running web testing and exploratory QA?

You need CDP-based browser access to run exploratory QA. The automation also utilizes persistent sessions, accessibility-tree snapshots, and authentication vault support to maintain reliable browser control during testing.

Can I capture screenshots during automated website navigation?

Yes, you can capture screenshots during website navigation. The automation workflow enables an AI agent to open pages, interact with interfaces, and take screenshots quickly without manual intervention.

How do I start automating workflows across Chrome and Chromium?

To start automating workflows across Chrome and Chromium, load the core workflow using the command line before running any tasks. This initializes the required AI control for subsequent browser automation actions.

What is the best way to handle form filling and logins without brittle manual steps?

The best way to handle form filling and logins is using AI agent control with persistent sessions and authentication vault support. This approach removes manual friction and ensures reliable automated interaction.