Parallel: Search - Deepline

Search the web for an objective or queries and get ranked pages with relevant excerpts, for repeated or large-scale lookups.

Tool ID: parallel_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(
  'parallel_search',
  {
    "mode": "one-shot",
    "objective": "Find recent fintech announcements and analyst reports.",
    "search_queries": [
      "software companies hiring engineers"
    ]
  },
);

console.log(result.toolResponse.raw);

CLI

deepline tools execute parallel_search --input '{
  "mode": "one-shot",
  "objective": "Find recent fintech announcements and analyst reports.",
  "search_queries": [
    "software companies hiring engineers"
  ]
}' --json

Example response

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

{
  "data": {
    "search_id": "search_123",
    "results": [
      {
        "url": "https://example.com"
      }
    ]
  }
}

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

Input reference

Parallel search input. Requires objective or search_queries.

Name	Type	Required	Default	Details
`payload.mode`	`"one-shot" \| "agentic"`	No	—	Allowed: `one-shot`, `agentic`.
`payload.objective`	`string`	No	—	Required unless search_queries is provided.
`payload.search_queries`	`array`	No	—	Required unless objective is provided.
`payload.max_results`	`integer`	No	—	Minimum: 1. Maximum: 20.
`payload.excerpts`	`object`	No	—	Parallel excerpt settings.
`payload.source_policy`	`object`	No	—	Parallel source policy.
`payload.fetch_policy`	`object`	No	—	Parallel fetch policy.
`payload.betas`	`string \| array`	No	—	—

Output reference

Parallel search result payload.

Name	Type	Required	Default	Details
`result.data`	`object`	Yes	—	Provider response payload.
`result.data.search_id`	`string`	Yes	—	—
`result.data.results`	`array`	Yes	—	—
`result.data.results[].url`	`string`	Yes	—	—
`result.data.results[].title`	`string \| null`	No	—	—
`result.data.results[].publish_date`	`string \| null`	No	—	—
`result.data.results[].excerpts`	`array \| null`	No	—	—
`result.data.warnings`	`array \| null`	No	—	—
`result.data.warnings[].type`	`string`	Yes	—	—
`result.data.warnings[].message`	`string`	Yes	—	—
`result.data.warnings[].detail`	`object \| null`	No	—	—
`result.data.session_id`	`string \| null`	No	—	—
`result.meta`	`record`	No	—	Additional response metadata (status, paging).

Deepline cost

Pricing model: per_result (per result).
Estimated Deepline credits: 0.02 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