Browse

Control a local headless browser to navigate, click, scroll, fill, and screenshot pages.

2|2|Updated Feb 12, 2026

Search Tags:#automation #playwright #browser #screenshot #browse #web-interaction

Ready to run

npx skills add https://github.com/bishwashere/cowCode --skill browse-bishwashere

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: Browse
Source: https://github.com/bishwashere/cowCode/tree/main/skills/browse
Command: npx skills add https://github.com/bishwashere/cowCode --skill browse-bishwashere

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Local automation of browser tasks reduces manual browsing time by controlling a local headless browser to navigate, click, fill forms, and capture proof without relying on cloud services.

Core Features & Use Cases

Persistent tab across messages so you can stay on the same page for follow-up actions.
Execute core browser actions (navigate, click, scroll, fill, screenshot) using selectors on the target site.
Use cases include opening product pages, gathering data, and generating proof (screenshots) of results for validation.

Quick Start

Navigate to a URL and capture a screenshot to prove the page content.

Frequently Asked Questions about Browse

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

FAQPage Schema

How do I automate browser tasks like clicking and scrolling on my local machine?▼

Local browser automation is achieved by running a headless Playwright (Chromium) instance to navigate, click, scroll, fill forms, and capture proof. It operates entirely on-device, using explicit selectors to interact with web pages without relying on cloud services.

Can I capture screenshots during web interaction to validate page content?▼

Yes, you can capture screenshots during web interaction to generate proof of page content. The local headless browser executes actions like navigation and form filling, then takes screenshots to validate results deterministically without using cloud services.

Does local browser automation work with persistent tabs across multiple messages?▼

Yes, local browser automation supports a persistent tab across multiple messages. This allows you to stay on the same page for follow-up actions, maintaining state as you navigate categories, click elements, and fill forms within the same headless session.

How do I navigate categories and fill forms using explicit selectors in Playwright?▼

To navigate categories and fill forms, you provide explicit selectors to the local headless browser. The automation requires these specific selectors to target elements on the website, handling interactions deterministically and returning proof screenshots.

What are the limitations of using a headless browser for local web interaction?▼

Limitations of local headless browser interaction include the requirement for explicit selectors to target elements, reducing flexibility with dynamic layouts. It also requires local Playwright setup and is constrained to on-device processing without cloud service integration.