What is the difference between web-search and scrape_url?

The web-search skill is used for discovering new pages based on a search query, while scrape_url is intended for extracting deep content from a specific, already-known URL.

How does the web-search skill handle bot detection?

It utilizes a headless browser with a specialized stealth configuration to navigate DuckDuckGo, which helps avoid CAPTCHAs and bot detection common with standard scrapers.

How many results can I expect from a single query?

The skill is configured to return the top 10 most relevant search results, including the title, URL, and a descriptive snippet for each.

Why is the first search slower than subsequent ones?

The first search takes approximately 5-6 seconds because it must launch and initialize the headless browser instance; subsequent searches use the existing instance and are faster.

Web Search

Name: Web Search
Author: cyrup-ai

bycyrup-ai

Web Scraping & Data Collection

Performs structured web searches via DuckDuckGo to retrieve real-time documentation, library information, and technical solutions.

About

The Web Search skill equips Claude with the ability to perform live internet searches, bridging the gap between its training data and the current state of the web. By utilizing a stealth-configured headless browser to query DuckDuckGo, it retrieves structured data—including page titles, URLs, and snippets—allowing Claude to research new libraries, find up-to-date documentation, and troubleshoot complex error messages using the latest community fixes. This skill is essential for developers who need to stay current with rapidly evolving frameworks and tools without leaving their terminal environment.

Key Features

High-relevance filtering returning the top 10 results per query
Support for technical research and library discovery workflows
Structured JSON output including rank, title, URL, and snippet
Stealth-configured DuckDuckGo searching to bypass bot detection
0 GitHub stars
Persistent browser sessions for faster subsequent search results

Use Cases

Discovering and evaluating third-party libraries or packages for specific project needs
Researching the latest best practices and syntax for modern programming frameworks
Searching for specific compiler error codes or stack traces to find community-driven solutions

About

Key Features

High-relevance filtering returning the top 10 results per query
Support for technical research and library discovery workflows
Structured JSON output including rank, title, URL, and snippet
Stealth-configured DuckDuckGo searching to bypass bot detection
0 GitHub stars
Persistent browser sessions for faster subsequent search results

Use Cases

Discovering and evaluating third-party libraries or packages for specific project needs
Researching the latest best practices and syntax for modern programming frameworks
Searching for specific compiler error codes or stack traces to find community-driven solutions