> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://mediakind.ferndocs.com/platform/getting-started/first-api-call/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://mediakind.ferndocs.com/_mcp/server.

# Your first API call

Before you build a larger integration, confirm your setup with one request. The goal is to make a single authenticated call against a project-scoped endpoint, inspect the response, and prove that your token and project access work.

## Prerequisites

You need:

* An MK.IO account with access to at least one project.

* A personal API token.

* The name of the project you want to query.

If you still need a token, follow [Authentication and tokens](/platform/getting-started/authentication).

## Set the values you reuse

Export the values the request needs, so the command stays readable and retries are less error-prone:

* Terminal window

```bash
export MKIO_BASE_URL="https://app.mk.io"
export MKIO_PROJECT="<PROJECT_NAME>"
export MKIO_TOKEN="<YOUR_TOKEN>"

```

## Send the request

Run the Media API list-assets operation:

```bash
curl -X GET "$MKIO_BASE_URL/api/v1/projects/$MKIO_PROJECT/media/assets" \
  -H "Authorization: Bearer $MKIO_TOKEN" \
  -H "Accept: application/json"

```

To see the HTTP status line as well, add `-i`:

```bash
curl -i -X GET "$MKIO_BASE_URL/api/v1/projects/$MKIO_PROJECT/media/assets" \
  -H "Authorization: Bearer $MKIO_TOKEN" \
  -H "Accept: application/json"

```

## Check the response

A successful response is HTTP `200` with a list body. List endpoints return two top-level fields:

`value`: the returned assets.

* `supplemental`: list metadata, including `supplemental.pagination` with the record counts.

```json
{
  "value": [
    { "name": "example-asset" }
  ],
  "supplemental": {
    "count": 1,
    "kind": "Asset",
    "operation": "list",
    "pagination": { "start": 0, "end": 1, "records": 1, "total": 1 }
  }
}

```

Many MK.IO list endpoints use this same shape, so it is worth recognising now. See [Pagination and filtering](/platform/understanding/pagination) for how to page and narrow these responses.

## If the call does not succeed

This endpoint returns the platform’s standard error responses:

| Status code | Meaning               | What to check next                                                     |
| ----------- | --------------------- | ---------------------------------------------------------------------- |
| `400`       | Bad Request           | Recheck the URL, query parameters, and request syntax.                 |
| `401`       | Unauthorized          | Recheck the bearer token and `Authorization` header.                   |
| `403`       | Forbidden             | Recheck that the user has access to the target project and operation.  |
| `404`       | Not Found             | Recheck the project name and path.                                     |
| `429`       | Too Many Requests     | Slow down and retry later.                                             |
| `500`       | Internal Server Error | Record the response `ref` value before retrying or contacting support. |

The error body carries `error.code` for programmatic handling, `error.detail` for logs, and `ref` for support follow-up. See [Error handling](/platform/understanding/error-handling) for the full model.

## What this confirms

A successful request proves that your token is valid, your base URL is correct, your project name is correct, and you can reach a real project-scoped endpoint. That is enough to move on to resource creation with far less guesswork.

## Next steps

* Which API to use: choose the API that owns your workflow.

* API overview: the patterns shared across the APIs.

* Build with the Media API: move into storage, assets, transforms, jobs, and publishing.