Web resources
Three components reach the web: http_request for APIs, web_scraper for page text, and web_search for finding pages. Each has its own reference page.
Works in workflow and agent mode.
| Resource | Use it for | Reference |
|---|---|---|
component: http_request | Call a JSON API, webhook, or REST endpoint | HTTP client |
component: web_scraper | Read the readable text of one page (title, headers, body, links) | Web scraper |
component: web_search | Find pages for a query (DuckDuckGo) | Search resources |
Which one
- Just data from an API - the
http_requestcomponent. Built-in auth, retry, and cache. - Text from a page you already have a URL for - the
web_scrapercomponent. No browser: the page is fetched and converted to markdown. - You need to find the right URL first - the
web_searchcomponent, thenweb_scraperon the result.
See also
- Authenticated API call tutorial -
http_requestauth, retry, cache - Search resources - find pages to fetch
- Resources overview - all resource types