Skip to Content
Agentic WorkspaceToolsBuilt-in toolsWeb

Web

Reach the open web — search, crawl pages, or call any HTTP API.

Search the web using your configured search provider. Category: Data Pull · Setup: none — works free out of the box.

web_search defaults to DuckDuckGo, which is free and needs no API key. Choosing a different provider — SerpAPI, Brave Search, Tavily, or Bing Search — requires adding that provider’s key under Configuration → Integrations.

ℹ️

Because those paid providers aren’t configured by default, web_search may show a Config Required badge in the registry — but it still works, using free DuckDuckGo. Add a provider key only if you want that specific engine.

  • Inputs: query (required); optional provider (default duckduckgo), num_results.
web_search tool — Overview tab
web_search tool — Test tab returning results via DuckDuckGo

web_crawler

Crawl web pages and extract clean text content. Category: Data Pull · Setup: none.

  • Inputs: a page url to crawl; returns extracted text.
web_crawler tool — Overview tab
web_crawler tool — Test tab

http_request

Make HTTP calls to any API — a client like Postman/Insomnia. Category: Data Pull · Setup: none (works without a key; add auth in the request if needed).

  • Inputs: method, url, headers, query params, body.
http_request tool — Overview tab
http_request tool — Test tab
💡

http_request is the same engine behind custom HTTP tools — use it for quick, ad-hoc calls; build a Project tool when you want a reusable, named endpoint.

UtilitiesDatabase