skills.rest
tonamson
tonamsonCOMMUNITY
Explore all skills by this author →
GitHub

ck:agent-browser

Automate browser tasks with snapshots, clicks, and state persistence via CLI.

Updated Aug 27, 2026
Search Tags:#cli#web-automation#browser-automation#session-persistence#cloud-browsers#autonomous-sessions
One-click install
npx skills add https://github.com/tonamson/please-done --skill ck-agent-browser-tonamson
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ck:agent-browser
Source: https://github.com/tonamson/please-done/tree/main/ck/.opencode/skills/agent-browser
Command: npx skills add https://github.com/tonamson/please-done --skill ck-agent-browser-tonamson

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

AI agents often need to automate complex browsing tasks while preserving context across long sessions. This Skill provides a CLI-driven solution that enables navigations, snapshots with element references, and reproducible interactions in a compact footprint.

Core Features & Use Cases

  • Open/Navigate: Launch and navigate to URLs with deterministic state.
  • Snapshot & Ref System: Capture interactive elements with refs for stable future interactions.
  • Actions & Automation: Perform clicks, fills, and checks, then re-snapshot to verify changes.
  • Context Persistence: Maintain sessions across steps and optionally operate with Browserbase cloud for CI/CD.
  • Use Case Example: Imagine QA teams running long user flows on a web app, recording each step and validating outcomes automatically.

Quick Start

Install agent-browser globally and begin a long autonomous browsing session to illustrate its capabilities.

Frequently Asked Questions about ck:agent-browser

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

FAQPage Schema
How do I maintain state during long autonomous browser automation sessions?▼

AI-driven browser automation uses a CLI with a snapshot and ref system to capture interactive elements and persist session state. This allows agents to execute long autonomous sessions by maintaining deterministic state across navigations, clicks, and fills without losing context.

How do I capture interactive web elements for stable future clicks and fills?▼

You capture interactive web elements using a snapshot and ref system that assigns stable references to page elements. This enables reproducible browser automation by allowing agents to target specific refs for future clicks, fills, and checks.

Can I run cloud browser automation testing across CI/CD pipelines?▼

Yes, you can run cloud browser automation testing across CI/CD pipelines by integrating with Browserbase cloud browsers. This enables self-verifying workflows and video recording of long user flows within your continuous integration environment.

Does browser automation work with self-verifying workflows for QA teams?▼

Browser automation supports self-verifying workflows for QA teams by performing actions and automatically re-snapshotting the page to validate changes. This allows teams to run long user flows, record each step, and verify outcomes automatically.

What is the best way to automate reproducible web interactions with a compact context footprint?▼

The best way to automate reproducible web interactions with a compact context footprint is using a CLI-driven solution that takes context-efficient snapshots. This approach captures necessary interactive elements with refs, ensuring deterministic state without overwhelming the agent's context.

Related Skills

Explore More Software Engineering Skills→
congthang12312congthang12312

agent-browser

Automate browser tasks via CLI with context-aware snapshots.

Community
Advanced
xixiUxixiU

testing

Automate end-to-end CI/CD QA pipelines from code commit to delivery.

Community
Advanced
NammDevNammDev
2

agent-browser

Automate browser tasks with snapshot refs and CLI commands.

Community
Advanced
sourcegraphsourcegraph
16

web-browser

Automate Chromium-based browser navigation, element interaction, and data extraction via CDP.

Official
Advanced