skills.rest
MinhQuyen274
MinhQuyen274COMMUNITY
Explore all skills by this author →
GitHub

ck:agent-browser

Automate browser navigation and interaction using snapshot-based @refs for AI agents.

Updated Aug 27, 2026
Search Tags:#browser automation#ai agents#form interaction#video recording#ci testing#snapshot refs#cloud browsers
One-click install
npx skills add https://github.com/MinhQuyen274/smartify --skill ck-agent-browser-minhquyen274
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ck:agent-browser
Source: https://github.com/MinhQuyen274/smartify/tree/main/Frontend_PROJECT_Figma/.opencode/skills/agent-browser
Command: npx skills add https://github.com/MinhQuyen274/smartify --skill ck-agent-browser-minhquyen274

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

It removes the friction of writing and maintaining complex browser automation scripts by letting AI agents navigate pages, identify interactive elements, and reuse stable element references across step-by-step interactions.

Core Features & Use Cases

  • Context-efficient “snapshot + refs” workflow: capture interactive elements into short @refs, then use those refs for subsequent actions reliably across page changes.
  • Full browser control for autonomous sessions: navigation, interaction (click/fill/type/drag), waits, media capture (screenshots/recording), and state persistence (save/load auth).
  • CI-friendly and cloud-browser support: run locally or on Browserbase for consistent headless testing and parallel sessions.

Quick Start

Run: agent-browser open https://example.com and then snapshot interactive elements to get refs for your next AI-driven actions.

Frequently Asked Questions about ck:agent-browser

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

FAQPage Schema
How do I automate browser interactions with AI agents using snapshot references?▼

AI agents automate browser navigation by capturing interactive elements into short @refs during a snapshot, then using those refs for subsequent actions like clicks and fills, and re-snapshotting after DOM changes to maintain reliable element tracking.

What is the snapshot and refs workflow for browser automation?▼

The snapshot and refs workflow captures interactive page elements into short @refs, allowing AI agents to execute actions like click, fill, or type using those stable refs. This approach solves element-finding and context-size problems by re-snapshotting after page changes.

Can I run browser automation testing in CI pipelines using cloud browsers?▼

Browser automation testing supports CI-friendly execution locally or on Browserbase cloud browsers. This enables consistent headless testing, parallel sessions, and video-enabled debugging for long autonomous workflows like self-verifying form interactions.

How do I handle DOM changes during autonomous browser workflows?▼

Handling DOM changes during autonomous browser workflows requires re-snapshotting interactive elements after page updates to generate fresh @refs. This snapshot-interact-re-snapshot cycle ensures AI agents maintain reliable element references throughout long sessions without breaking.

Does browser automation support session persistence and authentication state?▼

Browser automation supports session persistence by allowing you to save and load authentication state. This enables AI agents to maintain logged-in sessions across multiple navigation steps without re-authenticating during long autonomous workflows.

What are the limitations of using snapshot refs for AI browser automation?▼

Snapshot refs for AI browser automation require re-snapshotting after any DOM change to maintain valid references, adding overhead to dynamic pages. The approach depends on interactive elements being capturable, and complex shadow DOM or canvas content may pose challenges for element identification.

Related Skills

Explore More Software Engineering Skills→
DatTran26DatTran26

ck:agent-browser

Automates browser actions and manages sessions for AI agents and workflows via CLI commands.

Community
Advanced
AradotsoAradotso
4

kaboom-browser-ai-devtools

Correlate console logs, failed network requests, and DOM state into debugging evidence.

Official
Advanced
vm0-aivm0-ai
1.1k

agent-browser

Automate browser navigation, form filling, screenshots, and data extraction.

Official
Intermediate
kevinnguyen271090kevinnguyen271090
15

chrome-devtools

Automate browser tasks with Puppeteer CLI scripts for navigation, screenshots, and network monitoring.

Community
Advanced