---
title: Search
description: Search API endpoints.
---

## `dev.hatk.searchRecords`

Full-text search across a collection using SQLite FTS5 with BM25 ranking.

- **Type:** Query (GET)
- **Auth:** None

### Parameters

| Name         | Type    | Required | Default | Description               |
| ------------ | ------- | -------- | ------- | ------------------------- |
| `collection` | string  | Yes      | —       | Collection NSID to search |
| `q`          | string  | Yes      | —       | Search query              |
| `limit`      | integer | No       | `20`    | Results per page (1–100)  |
| `cursor`     | string  | No       | —       | Pagination cursor         |
| `fuzzy`      | boolean | No       | `true`  | Enable fuzzy matching     |

### Example

```bash
curl "http://127.0.0.1:3000/xrpc/dev.hatk.searchRecords?collection=fm.teal.alpha.feed.play&q=radiohead"
```

```typescript
import { callXrpc } from "$hatk/client";

const { items, cursor } = await callXrpc("dev.hatk.searchRecords", {
  collection: "fm.teal.alpha.feed.play",
  q: "radiohead",
});
```

### Response

```json
{
  "items": [ ... ],
  "cursor": "..."
}
```

### How it works

Hatk builds a SQLite FTS5 full-text search index for each collection. String fields in your record lexicon definitions are automatically included in the index, which uses incremental updates so new records become searchable immediately.

Search uses BM25 ranking to order results by relevance. The `fuzzy` parameter (enabled by default) allows approximate matching for typos and partial terms.