web-search

Search Google and Bing in real time via a Playwright-controlled browser.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/menhulu233/gearlai --skill web-search-menhulu233
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: web-search
Source: https://github.com/menhulu233/gearlai/tree/main/SKILLs/web-search
Command: npx skills add https://github.com/menhulu233/gearlai --skill web-search-menhulu233

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires express, playwright-core, uuid, and includes scripts (resource) components.

What problem does it solve?

Real-time information access and up-to-date data retrieval are often unavailable to Claude due to knowledge cutoffs. This Skill enables direct live web searches by driving a Playwright-controlled browser to fetch current content and present results in a structured Markdown format.

Core Features & Use Cases

  • Real-time web search across Google and Bing with automatic fallback to ensure results when one engine is unavailable.
  • Transparent, observable browser actions in a visible Chrome window during Cowork sessions.
  • Bridge-server API access for search, content retrieval, and page interactions, enabling automation and scripting.
  • Automatic lifecycle management integrated with GearlAI's Electron host to start and stop the service with the app.

Quick Start

Start the Bridge Server and then run the web-search CLI to perform your first live search.

Frequently Asked Questions about web-search

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

FAQPage Schema
How do I perform real-time web searches to bypass knowledge cutoffs in Claude?

Real-time web searches bypass knowledge cutoffs by driving a Playwright-controlled browser to fetch current content. This skill queries live information via Google and Bing, returning structured Markdown results directly to Claude.

Can I automate browser search and content retrieval using an API?

Browser search and content retrieval can be automated using a bridge-server API. It provides endpoints for executing searches, fetching page content, and performing page interactions with built-in fallbacks and error handling.

Does Playwright support visible Chrome browser automation during Cowork sessions?

Playwright supports visible Chrome browser automation during Cowork sessions. Browser actions remain transparent and observable in a visible window, allowing users to watch the automated search and retrieval process happen live.

How do I manage the lifecycle of a browser automation server in an Electron host?

Browser automation server lifecycle is managed automatically within an Electron host. The bridge server integrates with the app to auto-start and auto-stop the service, ensuring the search API is available only when needed.

What happens when a search engine fails during automated web search?

When a search engine fails during automated web search, the system implements automatic fallback. If Google or Bing is unavailable, it seamlessly switches to the alternative engine to ensure search results are consistently retrieved.

What is the best way to get up-to-date data retrieval formatted as Markdown?

Up-to-date data retrieval formatted as Markdown is best handled by querying live information through a controlled browser. This approach fetches current web content and synthesizes the results into a structured Markdown format for immediate use.