skills.rest
hidonguyen
hidonguyenCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

ck:agent-browser

Automate browser navigation and element interaction via refs-based CLI snapshots.

Updated Apr 2, 2026
Search Tags:#cli#playwright#session-management#snapshot#browser-automation#video-recording#browserbase
One-click install
npx skills add https://github.com/hidonguyen/trade-ops --skill ck-agent-browser-hidonguyen
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ck:agent-browser
Source: https://github.com/hidonguyen/trade-ops/tree/main/.opencode/skills/agent-browser
Command: npx skills add https://github.com/hidonguyen/trade-ops --skill ck-agent-browser-hidonguyen

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Agent-browser removes the heavy contextual overhead and fragility of traditional browser automation by providing compact, refs-based snapshots and a CLI optimized for long-running autonomous sessions, video recording, and cloud browser execution.

Core Features & Use Cases

  • Compact snapshots & refs: Capture interactive elements with short refs (e.g., @e1) to minimize token usage in AI workflows.
  • Deterministic CLI interactions: Navigate, fill, click, and assert using a stable command set suitable for CI/CD and autonomous agents.
  • Media & debugging: Built-in screenshot, PDF export, and video recording for reproducible debugging and audit trails.
  • Cloud & session support: Integrates with Browserbase and named sessions for parallel and remote execution.
  • Use Case: Run an end-to-end automated login and checkout flow in CI using cloud browsers while saving authenticated state and a video recording for verification.

Quick Start

Use ck:agent-browser to open a webpage, capture an interactive snapshot, interact with elements by their refs, and save a screenshot or recording for debugging.

Frequently Asked Questions about ck:agent-browser

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

FAQPage Schema
How do I reduce token usage when running long browser automation tasks with AI agents?▼

Browser automation for long-running AI sessions uses compact, refs-based snapshots to minimize token usage. Interactive elements are captured with short references like @e1, enabling efficient autonomous agent workflows without heavy contextual overhead.

Does Browserbase work with Playwright for remote browser execution in CI/CD?▼

Yes, Browserbase integration supports cloud browser execution and named sessions for parallel remote workflows. You can run deterministic Playwright-based CLI interactions, navigate, and assert in CI/CD environments using cloud browsers.

How do I capture screenshots and video recordings during a browser automation session?▼

Built-in media and debugging features capture screenshots, export PDFs, and record videos during browser automation sessions. These outputs provide reproducible debugging and audit trails for autonomous agent workflows and CI/CD browser tests.

Can I persist authenticated state across multiple browser automation sessions?▼

Yes, session persistence and named sessions allow you to save authenticated state across browser automation runs. This enables long-running AI sessions and CI/CD browser tests to maintain login states without repeated authentication.

What is the best way to interact with web elements using a CLI for autonomous agents?▼

Deterministic CLI interactions let you navigate, fill, click, and assert using stable commands and short element refs like @e1. This approach is optimized for autonomous agents and produces JSON-friendly CLI outputs for seamless integration.

Why does traditional browser automation fail in context-constrained AI workflows?▼

Traditional browser automation fails in context-constrained AI workflows due to heavy contextual overhead and fragility from large DOM snapshots. Compact, refs-based snapshots solve this by capturing only interactive elements with short references, minimizing token usage.

Related Skills

Explore More Software Engineering Skills→
pengfitpengfit

playwright-cli

Execute Playwright CLI commands for browser automation and web scraping.

Community
Intermediate
abnejsolutions-altabnejsolutions-alt

evidence-artifact-collector

Collects point-in-time audit evidence from AWS, Azure, and GCP into JSON/Markdown reports.

Community
Intermediate
NammDevNammDev
2

agent-browser

Automate browser tasks with snapshot refs and CLI commands.

Community
Advanced
lackeyjblackeyjb
3.0k

Playwright Browser Automation

Automate browser testing and interactions with Playwright scripts.

Community
Intermediate