browse

Automate web browsing for Claude Code workflows via a headless Chromium daemon.

1|Updated Mar 20, 2026
One-click install
npx skills add https://github.com/0xHarbs/agent-setup --skill browse-0xharbs
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: browse
Source: https://github.com/0xHarbs/agent-setup/tree/main/claude/skills/automation/browse
Command: npx skills add https://github.com/0xHarbs/agent-setup --skill browse-0xharbs

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Claude Code users need reliable, autonomous web access to verify deployments, read dynamic content, and interact with web pages without manual browser handling. This skill provides a persistent headless browser service that translates natural-language prompts into deterministic, scriptable web actions.

Core Features & Use Cases

  • Persistent headless Chromium daemon for fast, repeatable web interactions.
  • CLI-style command surface to navigate, read, fill forms, run JS, and capture logs.
  • Use Case: QA engineers can quickly validate deployment URLs by automating navigation, content extraction, and visual checks in a single workflow.

Quick Start

Launch the persistent browser and issue a few commands to navigate pages, capture data, and verify deployments.

Frequently Asked Questions about browse

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

FAQPage Schema
How do I automate headless browser interactions within Claude Code workflows?

Automate headless browser interactions by using a persistent headless Chromium daemon that translates natural-language prompts into deterministic web actions via a CLI-style command surface. This enables navigation, form filling, and content extraction directly inside Claude Code workflows.

Can I use Playwright to verify deployments and read dynamic web pages autonomously?

Yes, you can verify deployments and read dynamic pages using a persistent headless Chromium daemon that captures logs, runs JavaScript, and extracts content autonomously without manual browser handling.

What's the best way to perform QA automation on dynamic sites without manual browser handling?

Use a persistent headless browser service with a token-authenticated HTTP API and ref-based element snapshot to drive deterministic QA automation interactions, enabling repeatable content extraction and deployment validation.

Does this headless browser tool support running JavaScript and filling forms on dynamic pages?

Yes, the headless browser tool supports running JavaScript, filling forms, navigating pages, and capturing logs through its CLI-style command surface designed for deterministic interactions with dynamic web pages.

How does the ref-based element snapshot mechanism work for web automation?

The ref-based element snapshot mechanism captures page elements as deterministic references, allowing the headless browser to drive reliable interactions by targeting specific elements through a token-authenticated HTTP API.

What are the limitations of using a persistent headless Chromium daemon for web browsing automation?

The persistent headless Chromium daemon is designed for deterministic web automation tasks like verifying deployments and reading docs, but may require careful token-authenticated API management and is limited to headless browser interactions without visual rendering.