agent-browser

Automate Chromium-based browser tasks via CDP for navigation, form filling, and screenshots.

Updated Apr 26, 2026
One-click install
npx skills add https://github.com/suwandre/my-opencode-config --skill agent-browser-suwandre
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-browser
Source: https://github.com/suwandre/my-opencode-config/tree/main/skills/agent-browser
Command: npx skills add https://github.com/suwandre/my-opencode-config --skill agent-browser-suwandre

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Manual web interactions are repetitive, error-prone, and hard to reproduce across environments. This skill provides a deterministic, scriptable way to automate browser tasks for AI agents.

Core Features & Use Cases

  • Deterministic browser automation: drive pages, fill forms, click controls, take screenshots, and extract data.
  • Testing and QA automation: validate web app flows, logins, and navigation across environments.
  • Use Case: Automate routine web tasks for data collection and validation in AI-assisted development workflows.

Quick Start

Open a target URL, snapshot the page to capture element refs, and perform a click or form fill to complete an 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 web browser tasks for AI agents?

Automate web browser tasks for AI agents by controlling Chromium-based browsers through CDP. This skill provides deterministic, scriptable commands for navigation, form filling, click actions, and data extraction across websites and web apps.

How does browser automation through CDP handle authentication flows?

Browser automation through CDP supports named sessions to manage authentication flows. This allows you to maintain persistent login states and execute scriptable commands like open, snapshot, fill, and click across deterministic web testing sequences.

What is the best way to extract data from web pages without manual interaction?

The best way to extract data from web pages without manual interaction is using deterministic browser automation. By controlling Chromium-based browsers via CDP, you can script navigation, capture snapshots for element refs, and extract data reproducibly across environments.

Can I use CLI commands for web testing and QA automation?

Yes, you can use CLI commands for web testing and QA automation. This approach provides deterministic commands to validate web app flows, logins, and navigation, enabling reproducible testing across different environments without manual browser interaction.

Do I need to install external dependencies to control Chromium browsers via CDP?

No external dependencies are required to control Chromium browsers via CDP. The skill operates independently to provide deterministic, scriptable commands for navigation, form filling, and screenshot capture directly through its browser CLI interface.

Why use deterministic browser automation over manual web interaction?

Deterministic browser automation replaces manual web interaction because manual tasks are repetitive, error-prone, and hard to reproduce. Scriptable CDP commands ensure consistent navigation, form filling, and data extraction across various environments for reliable AI-assisted workflows.