Brave Search

• You need to search for documentation, API references, or technical facts from the web.
• You want to fetch and read the content of a specific URL as clean markdown.
• You need web search results in a lightweight, scriptable way without a full browser or Playwright dependency.
• Your agent task requires current information or web lookup as a sub-step.

• You need to interact with a web page (click buttons, fill forms, log in) — use a browser-automation skill instead.
• You need real-time data from sources that block crawlers or require JavaScript rendering not handled by simple fetch.

Queries the Brave Search API and returns up to 10 results per request, each with title, link, and snippet. No browser process is spawned, keeping it fast and resource-efficient.

Use the --content flag to fetch and extract readable content from each result page, returned as clean markdown. Ideal for summarization, Q&A, or deeper research tasks.

The content.js command fetches any URL directly and converts it to readable markdown, letting your agent pull structured content from a known page without a search step.

Control how many results are returned with the -n flag, from a quick 5-result lookup up to 10 results for broader coverage.

Search for API references or library docs by query, optionally fetching full page content so your agent can answer questions with up-to-date technical details.

Retrieve current facts or news snippets from the web to ground agent responses in real-world information without maintaining a browser session.

Point the extractor at a specific article, changelog, or reference page to get its full text as markdown, ready for summarization or analysis.