Search the web with Firecrawl

searchWebWithFirecrawl searches one or more Firecrawl source families. Add scrapeOptions when you need page content instead of result metadata alone. Inputs accept literal values or compatible signals. An optional second argument { account } selects the Firecrawl binding.

Example

automations/search-product-news.automation.ts

import { automation, onSchedule } from "automate.ax"
import { searchWebWithFirecrawl } from "automate.ax/firecrawl"

export default automation("Search product news", () => {
  onSchedule({ schedule: "0 9 * * 1" })

  searchWebWithFirecrawl({
    query: '"Automate.ax" release',
    sources: ["web", "news"],
    tbs: "qdr:w",
    limit: 10,
    scrapeOptions: { formats: ["markdown", "summary"] },
  })
})

Inputs

Input	Type	Required	Default	Description
`query`	`string`	Yes	—	Search query, up to 500 characters. Firecrawl supports operators such as quotes, `site:`, `filetype:`, and exclusions prefixed with `-`.
`sources`	`('web' \| 'news' \| 'images' \| { type })[]`	No	`['web']`	Result families to request. Object entries use the same source names.
`categories`	`('github' \| 'research' \| 'pdf' \| 'developer' \| { type })[]`	No	None	Narrow ordinary web results or add developer results. Object entries use the same category names. `research` filters academic websites; it is not Firecrawl’s research-paper index.
`limit`	`number`	No	`10`	Maximum results per source, from 1 through 100.
`includeDomains`	`string[]`	No	None	Restrict results to these hostnames. Cannot be combined with `excludeDomains`.
`excludeDomains`	`string[]`	No	None	Exclude these hostnames. Cannot be combined with `includeDomains`.
`tbs`	`string`	No	None	Firecrawl time filter, such as `qdr:h`, `qdr:d`, `qdr:w`, `qdr:m`, `qdr:y`, or a custom date range.
`location`	`string`	No	None	Geographic search location.
`ignoreInvalidURLs`	`boolean`	No	`false`	Remove results that cannot be passed to other Firecrawl endpoints.
`highlights`	`boolean`	No	Provider default	Generate query-relevant result highlights.
`timeout`	`number`	No	`60000`	Maximum provider search time in milliseconds.
`scrapeOptions`	`object`	No	None	The same formats, browser actions, content filters, cache, proxy, redaction, and target settings accepted by `scrapeUrlWithFirecrawl`.
`enterprise`	`('default' \| 'anon' \| 'zdr')[]`	No	None	Enterprise privacy modes enabled for the Firecrawl team.
`threatProtection`	`object`	No	Account policy	Enterprise threat-protection overrides.

Output

Returns Firecrawl’s optional web, news, images, and developer arrays. Firecrawl omits buckets it did not return. Web and developer results contain web fields such as url, title, description, position, and category; news and image results retain their own snippet, date, imageUrl, and dimension fields. When scrapeOptions is present, results can also contain the document fields returned by scrapeUrlWithFirecrawl.

Get started

Guides

Concepts

Management API

Reference

Search the web with Firecrawl

Example

Inputs

Output

​Example

​Inputs

​Output

Example

Inputs

Output