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

# Headers

> Content-Type, Content-Encoding, Accept-Encoding, and Meili-Include-Metadata headers for the Meilisearch API.

## Content type

Any API request with a payload (`--data-binary`) requires a `Content-Type` header. Content type headers indicate the media type of the resource, helping the client process the response body correctly.

Meilisearch currently supports the following formats:

* `Content-Type: application/json` for JSON
* `Content-Type: application/x-ndjson` for NDJSON
* `Content-Type: text/csv` for CSV

Only the [add documents](/reference/api/documents#add-or-replace-documents) and [update documents](/reference/api/documents#add-or-update-documents) endpoints accept NDJSON and CSV. For all others, use `Content-Type: application/json`.

## Content encoding

The `Content-Encoding` header indicates the media type is compressed by a given algorithm. Compression improves transfer speed and reduces bandwidth consumption by sending and receiving smaller payloads. The `Accept-Encoding` header, instead, indicates the compression algorithm the client understands.

Meilisearch supports the following compression methods:

* `br`: uses the [Brotli](https://en.wikipedia.org/wiki/Brotli) algorithm
* `deflate`: uses the [zlib](https://en.wikipedia.org/wiki/Zlib) structure with the [deflate](https://en.wikipedia.org/wiki/DEFLATE) compression algorithm
* `gzip`: uses the [gzip](https://en.wikipedia.org/wiki/Gzip) algorithm

### Request compression

The code sample below uses the `Content-Encoding: gzip` header, indicating that the request body is compressed using the `gzip` algorithm:

```
 cat ~/movies.json | gzip | curl -X POST 'MEILISEARCH_URL/indexes/movies/documents' --data-binary @- -H 'Content-Type: application/json' -H 'Content-Encoding: gzip'
```

### Response compression

Meilisearch compresses a response if the request contains the `Accept-Encoding` header. The code sample below uses the `gzip` algorithm:

```
curl -sH 'Accept-encoding: gzip' 'MEILISEARCH_URL/indexes/movies/search' | gzip -dc
```

## Search metadata

You may use an optional `Meili-Include-Metadata` header when performing search and multi-search requests:

```
curl -X POST 'http://localhost:7700/indexes/INDEX_NAME/search' \
  -H 'Content-Type: application/json' \
  -H 'Authorization: Bearer MEILISEARCH_KEY' \
  -H 'Meili-Include-Metadata: true' \
  -d '{"q": ""}'
```

Meilisearch Cloud includes this header by default.

Responses will include a `metadata` object:

```json theme={null}
{
  "hits": [ … ],
  "metadata": {
    "queryUid": "0199a41a-8186-70b3-b6e1-90e8cb582f35",
    "indexUid": "INDEX_NAME",
    "primaryKey": "INDEX_PRIMARY_KEY"
  }
}
```

`metadata` contains the following fields:

|     Field    |   Type  |                         Description                        |
| :----------: | :-----: | :--------------------------------------------------------: |
|  `queryUid`  | UUID v7 |               Unique identifier for the query              |
|  `indexUid`  |  String |                      Index identifier                      |
| `primaryKey` |  String |     Primary key field name, if index has a primary key     |
|   `remote`   |  String | Remote instance name, if request targets a remote instance |

<Note>
  A search refers to a single HTTP search request. Every search request is assigned a `requestUid`. A query UID is a combination of `q` and `indexUid`.

  In the context of multi-search, for any given `searchUid` there may be multiple `queryUid` values.
</Note>