Web search and content extraction via Brave Search API. Use for searching documentation, facts, or any web content. Lightweight, no browser required.
The 'brave-search' skill provides a headless web search and content extraction tool that leverages the Brave Search API. It solves the problem of needing to access web content without requiring a browser, offering a lightweight solution that performs searches and extracts relevant documentation, facts, or web-based information. This makes it ideal for automated workflows, bots, or other tasks where traditional browser-based search isn't an option or is too resource-intensive.
To set up the skill, navigate to the brave-search directory and run `npm ci` to install dependencies. Additionally, make sure the `BRAVE_API_KEY` environment variable is set.
Yes, you can extract content from specific URLs using the `./content.js` script. Simply provide the URL to fetch and extract readable content as markdown.
The output includes the title, link, a snippet from the search results, and the page content (if the `--content` flag is used). The content is provided in markdown format.
This skill relies on the Brave Search API and is limited to its capabilities. It does not support interactive browsing and can only return up to the specified number of search results.
The skill supports basic search queries, where you can specify the number of results or include page content. You can search for documentation, facts, or content from specific URLs.
Quick Setup:
.claude/skills/Repository
steipete/agent-scripts