ck:agent-browser

Automate browser tasks via a refs-based CLI for long-running sessions.

1|Updated Mar 5, 2026
One-click install
npx skills add https://github.com/hotriluan/alkana_web --skill ck-agent-browser-hotriluan
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ck:agent-browser
Source: https://github.com/hotriluan/alkana_web/tree/main/.opencode/skills/agent-browser
Command: npx skills add https://github.com/hotriluan/alkana_web --skill ck-agent-browser-hotriluan

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

AI agents often struggle with browser tasks due to heavy context requirements and brittle flows. This Skill provides a lightweight, context-efficient CLI that uses a refs-based interaction model to streamline long autonomous sessions, enable self-verifying workflows, and support video recording and cloud browser testing.

Core Features & Use Cases

  • Refs-based interactions: snapshot, interact, and verify elements with short context.
  • Long-running sessions and self-verification: maintain progress across steps and recover from interruptions.
  • Cloud/browser testing: integrate with Browserbase for scalable, remote execution and CI/CD.

Quick Start

Install agent-browser globally and start a session to navigate, snapshot, and interact with web pages.

Frequently Asked Questions about ck:agent-browser

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

FAQPage Schema
How do I run AI-powered browser automation for long autonomous sessions without exceeding context limits?

AI-powered browser automation for long sessions uses a refs-based interaction model to snapshot and interact with elements using short context, minimizing token usage and maintaining progress across steps.

What's the best way to automate self-verifying browser workflows that recover from interruptions?

Self-verifying browser workflows use a refs-based CLI interaction model to maintain progress across steps and recover from interruptions, enabling autonomous agents to verify element states without brittle flows.

Can I integrate AI browser automation with Browserbase for cloud testing and CI/CD?

AI browser automation integrates with Browserbase to support scalable remote execution, video recording, and cloud browser testing within CI/CD pipelines for autonomous sessions.

Why do AI agents struggle with browser tasks and how does refs-based interaction solve it?

AI agents struggle with browser tasks due to heavy context requirements and brittle flows. A refs-based interaction model solves this by using short context snapshots to interact with and verify elements efficiently.

How do I set up a CLI browser automation environment for autonomous web testing?

Setting up CLI browser automation requires a root SKILL.md with YAML name and description metadata, optionally including scripts, references, and assets directories to configure extended autonomous testing capabilities.

Does agent-browser support video recording during long-running autonomous web sessions?

Agent-browser supports video recording and cloud browser testing during long-running autonomous sessions through Browserbase integration, enabling scalable remote execution and session playback.