brave-search

Web content extraction as Markdown is performed by querying the Brave Search API and parsing retrieved pages using Readability and Turndown. The extracted main text is formatted directly as Markdown, streamlining information gathering without a browser.

To perform programmatic web searches, you need a valid BRAVE_API_KEY environment variable and specific npm dependencies installed. These prerequisites allow the scripts to query the Brave Search API and extract page content.

Retrieving API documentation directly within your coding workflow is supported by executing targeted web searches and extracting page content. The Skill queries the Brave Search API to fetch factual information and returns concise summaries or full content snippets.

Web scraping with Readability and jsdom works for extracting main page content by evaluating the Document Object Model. Readability isolates the primary text, while jsdom provides the environment to parse the HTML structure before conversion.

Limitations of using the Brave Search API for web scraping include the strict requirement of a valid BRAVE_API_KEY and reliance on npm dependencies. If the target page lacks a clear text structure, Readability may fail to extract the main content effectively.

The best way to convert extracted HTML content to GitHub Flavored Markdown is by using Turndown alongside the turndown-plugin-gfm. This combination translates the parsed HTML from jsdom into structured Markdown for programmatic information retrieval.