Exa: Search - Deepline

Semantic web search for pages, companies, people, LinkedIn profiles, news and sources, with date and domain filters.

Tool ID: exa_search

Run this action

Use the TypeScript SDK for one call. Use ctx.tools.execute(...) inside a Play for durable calls, scheduled calls, or CSV processing.

import { Deepline } from 'deepline';

const deepline = await Deepline.connect();
const result = await deepline.tools.execute(
  'exa_search',
  {
    "query": "software companies hiring engineers"
  },
);

console.log(result.toolResponse.raw);

CLI

deepline tools execute exa_search --input '{
  "query": "software companies hiring engineers"
}' --json

Example response

The SDK returns this shape at result.toolResponse.raw. The example values illustrate the response.

{
  "data": {
    "requestId": "request_123",
    "results": [
      {
        "id": "id_123",
        "url": "https://example.com",
        "title": "VP of Engineering"
      }
    ],
    "context": "example"
  }
}

Use deepline describe exa_search --json for the latest machine-readable contract.

Input reference

Exa raw search input.

Name	Type	Required	Default	Details
`payload.query`	`string`	Yes	—	The search query string.
`payload.additionalQueries`	`array`	No	—	Additional queries (used with type=“deep” to expand coverage).
`payload.type`	`"auto" \| "fast" \| "deep" \| "neural" \| "instant"`	No	—	Search type (auto chooses the best strategy). Allowed: `auto`, `fast`, `deep`, `neural`, `instant`.
`payload.category`	`"company" \| "people" \| "news" \| "research paper" \| "tweet" \| "personal site" \| "financial report"`	No	—	Optional result category filter. Allowed: `company`, `people`, `news`, `research paper`, `tweet`, `personal site`, `financial report`.
`payload.numResults`	`number`	No	—	Number of results to return (max 100).
`payload.includeDomains`	`array`	No	—	Only include results from these domains.
`payload.excludeDomains`	`array`	No	—	Exclude results from these domains.
`payload.startCrawlDate`	`string`	No	—	Only include links crawled after this ISO date-time.
`payload.endCrawlDate`	`string`	No	—	Only include links crawled before this ISO date-time.
`payload.startPublishedDate`	`string`	No	—	Only include links published after this ISO date-time.
`payload.endPublishedDate`	`string`	No	—	Only include links published before this ISO date-time.
`payload.includeText`	`array`	No	—	Require these text snippets to appear in page content (max 1 string of up to 5 words). Maximum items: 1.
`payload.excludeText`	`array`	No	—	Exclude pages containing these text snippets (max 1 string of up to 5 words). Maximum items: 1.
`payload.userLocation`	`string`	No	—	Two-letter ISO country code, e.g. “US”.
`payload.contents`	`object`	No	—	—
`payload.context`	`boolean \| object`	No	—	—
`payload.moderation`	`boolean`	No	—	Enable content moderation (may reduce recall).

Output reference

Exa search result.

Name	Type	Required	Default	Details
`result.data`	`object`	Yes	—	Provider response payload.
`result.data.requestId`	`string`	No	—	—
`result.data.results`	`array`	No	—	—
`result.data.results[].id`	`string`	No	—	—
`result.data.results[].url`	`string`	No	—	—
`result.data.results[].title`	`string`	No	—	—
`result.data.results[].text`	`string`	No	—	—
`result.data.results[].publishedDate`	`string`	No	—	—
`result.data.results[].author`	`string \| unknown`	No	—	—
`result.data.results[].score`	`number`	No	—	—
`result.data.results[].highlights`	`array`	No	—	—
`result.data.results[].highlightScores`	`array`	No	—	—
`result.data.results[].summary`	`string \| record`	No	—	—
`result.data.results[].entities`	`array`	No	—	—
`result.data.results[].entities[].id`	`string`	No	—	—
`result.data.results[].entities[].type`	`"company" \| "person"`	No	—	Allowed: `company`, `person`.
`result.data.results[].entities[].version`	`number`	No	—	—
`result.data.results[].entities[].properties`	`record`	No	—	—
`result.data.results[].extras`	`record`	No	—	—
`result.data.context`	`string`	No	—	—
`result.meta`	`record`	No	—	Additional response metadata (status, paging).

Deepline cost

Pricing model: per_result (per result).
Estimated Deepline credits: 0.1 per pricing unit.

​Run this action

​CLI

​Example response

​Input reference

​Output reference

​Deepline cost

​Related documentation

Run this action

CLI

Example response

Input reference

Output reference

Deepline cost

Related documentation