> ## Documentation Index
> Fetch the complete documentation index at: /docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Company research

> Find each company's news stories from a list of domains.

Send `companies` to `POST /runs`. The agent finds each company's news stories
in the window you set, optionally only the ones about a topic.

```json theme={null}
{
  "companies": {
    "items": [
      { "domain": "stripe.com" },
      { "domain": "adyen.com", "name": "Adyen" }
    ],
    "save_as_list": true,
    "list_name": "Payments"
  },
  "query": "funding and acquisitions",
  "lookback_days": 90
}
```

## Parameters

**companies** `object` *required*

The companies: `list_id` of a [company list](/docs/agent/company-lists.md), or
`items`, each with a `domain` (a URL works too) and an optional `name`.
`save_as_list: true` also saves typed items as a list, named `list_name`.

**query** `string`

Keep only stories about this, for example "funding and acquisitions". Blank:
every story about the companies.

**lookback\_days** `integer` (default: 30)

How many days back from today: 1 to 365.

**include\_market\_activity** `boolean` (default: true)

Keep market activity stories: stock moves and analyst ratings.

**read\_web** `boolean` (default: true)

Also read each company's own newsroom and the web, not only the news.

**recheck\_borderline** `boolean` (default: true)

A stronger model rechecks borderline web pages for companies with little
coverage.

**keep\_updated** `object`

Make a company monitor of the run once it succeeds. See
[Monitors](/docs/agent/monitors.md).

Company runs always use the `standard` mode.

## Large lists

Free and pay-as-you-go runs take up to 50 companies. On
[Enterprise plans](/docs/get-started/enterprise-plans.md), a run of more than 50
companies is split into runs of up to 50, grouped in a
[batch](/docs/agent/runs.md#batches). The response then has a `batch_id` and the
`run_ids` of every part.

## Coverage per company

Besides the stories, a companies run returns one row per company in
`companies`, with what the agent found:

| Status | Meaning |
| - | - |
| `found` | Stories were found. |
| `no_news` | The company was searched and has no stories in the window. |
| `incomplete` | The search for the company did not finish. |
| `identity_unknown` | The agent could not tell which company the domain is. |
| `pending` | Not reached yet; the run is still going. |
| `not_searched` | The run ended before reaching the company. |

Download it with `GET /runs/{run_id}/companies.csv` or `companies.xlsx`.

If the run used a company list, fix a company with `identity_unknown` in the
list: see [Fix a company](/docs/agent/company-lists.md#fix-a-company).


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.