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

# Filter reference

> Every value search_jobs accepts, and what happens when you pass something else.

`search_jobs` validates every filter against the same service the website uses.
Passing an unrecognised value returns an error listing what is allowed, so an
assistant can correct itself without guessing.

## ATS source

`platform` — which applicant tracking system the role was posted through.

| Value         | Source                   |
| ------------- | ------------------------ |
| `greenhouse`  | Greenhouse job boards    |
| `ashby`       | Ashby job boards         |
| `ycombinator` | Y Combinator's job board |

## Employment type

`employment_type`

`Full-Time` · `Part-Time` · `Contract`

<Note>
  Values are case-sensitive and hyphenated exactly as shown.
</Note>

## Experience level

`experience_level`

`Intern` · `Junior` · `Mid-level` · `Senior` · `Staff` · `Lead` · `Manager` · `Director`

## Recency

`posted` — only roles first seen within the window.

| Value   | Window               |
| ------- | -------------------- |
| `today` | Since midnight today |
| `3d`    | Last 3 days          |
| `7d`    | Last week            |
| `30d`   | Last month           |

## Location

Two independent filters, usable together:

<ParamField body="location" type="string">
  Substring match against the role's location text — `Remote`, `Berlin`,
  `New York`. Good for free-form matching, including remote roles.
</ParamField>

<ParamField body="country" type="string">
  ISO alpha-3 country code — `USA`, `IND`, `DEU`, `GBR`. Matched on word
  boundaries, so `IND` will not also match "Indianapolis".
</ParamField>

## Domain

`domain` — a slug from `list_domains`, e.g. `software-engineering`,
`ai-data-science`, `design`.

<Tip>
  Call `list_domains` rather than guessing. It returns all 27 domains with their
  current open-role counts, and the slugs are the authoritative values.
</Tip>

## Company

`company` — a company slug, to restrict results to a single employer. For a
fuller picture of one company, `get_company_overview` is usually the better tool.

## Skill

`skill` — a skill slug, e.g. `python`, `kubernetes`, `react`. Matches jobs that
list that skill among their requirements.

<Tip>
  `search_jobs` results don't carry the full description, so a skill isn't
  always visible in the requirements text a listing shows — it only reflects
  what the ingestion pipeline extracted as a structured skill. If you need the
  full JD to check for a skill mentioned in prose, call `get_job_details`.
</Tip>

## Pagination

<ParamField body="page" type="integer">
  1-based page number. Defaults to `1`.
</ParamField>

<ParamField body="limit" type="integer">
  Results per page, between 1 and 50. Defaults to `20`.
</ParamField>

Responses include `total`, `page` and `total_pages` so an assistant knows whether
more results exist.

## Combining filters

Filters combine with **AND**. This finds senior or staff remote engineering roles
posted in the last week through Ashby:

```json theme={null}
{
  "domain": "software-engineering",
  "experience_level": "Senior",
  "location": "Remote",
  "platform": "ashby",
  "posted": "7d"
}
```

<Warning>
  Stacking many filters at once can return zero results. Narrow gradually —
  start broad, then add constraints — so it stays clear which filter emptied the
  set.
</Warning>