Photos

Any step can capture photos in addition to its primary answer. Photos are uploaded directly from the device to Rundun's cloud storage — the API never handles photo bytes.

Photo overlay

To enable the camera on any step, set the photo field to an object:

{
  "id": "s-cl01",
  "type": "boolean",
  "label": "Is the unit clean and ready?",
  "required": true,
  "photo": { "min": 1, "max": 4, "annotate": false },
  "config": { "true_label": "Yes", "false_label": "No" }
}

The executor must take at least min photos before the step can be completed. The camera button appears after the primary answer is given.

Setting photo: false or omitting photo entirely means no camera is shown for that step (default).

Photo step type

Use type: "photo" when the primary answer itself is one or more photos — no other answer type is involved:

{
  "id": "s-fp01",
  "type": "photo",
  "label": "Front of vehicle",
  "hint": "Take a straight-on shot from 3 metres away",
  "required": true,
  "photo": false,
  "config": { "min": 1, "max": 3, "annotate": false }
}

For photo type steps, the photo overlay field is false — the config on the step itself controls the min/max. The answer value in webhook payloads is null; the photos are in the photos array.

Upload flow

Photos use a two-phase upload to keep the API fast and the webhook payload lightweight:

1. On step completion: The device compresses the photo to a 320px wide JPEG thumbnail (~25KB each) and sends it inline in the API request that records the step answer. This thumbnail is immediately embedded in the step.completed webhook — no separate fetch required.

2. Full-resolution upload: Simultaneously, the device requests a pre-signed upload URL from the API:

POST /v1/runs/:run_id/steps/:step_id/photos/upload-url

The device uploads the full-resolution photo directly to Rundun's cloud storage using the pre-signed URL. The API is not in the upload path — it never handles photo bytes.

3. After run completion: The run.completed webhook includes full-resolution photo URLs instead of base64 thumbnails. By the time the run is submitted, the background uploads are typically complete.

Webhook payload shapes

In step.completed — base64 thumbnail

Available immediately when the step is answered. Approximately 25KB per photo.

"photos": [
  {
    "photo_index": 0,
    "preview": "data:image/jpeg;base64,/9j/4AAQSkZJRgABAQAAAQABAAD...",
    "width": 320,
    "height": 240
  },
  {
    "photo_index": 1,
    "preview": "data:image/jpeg;base64,/9j/4AAQSkZJRgABAQAAAQABAAD...",
    "width": 320,
    "height": 240
  }
]

A step with 5 photos adds approximately 125KB to the step.completed payload — acceptable for live progress tracking.

In run.completed — full-resolution media URL

Full-resolution photos uploaded from the device. At 25–50 photos per run, inline base64 would exceed 1.25MB which most receivers reject — so run.completed uses URLs instead.

"photos": [
  {
    "photo_index": 0,
    "url": "https://media.rundun.app/runs/r-550e8400-e29b-41d4-a716-446655440000/s-dm02/photo_0.jpg"
  },
  {
    "photo_index": 1,
    "url": "https://media.rundun.app/runs/r-550e8400-e29b-41d4-a716-446655440000/s-dm02/photo_1.jpg"
  }
]

URLs are stable and publicly accessible for the duration of your plan's storage retention period.

Storage

Photos are stored in Rundun's cloud storage at media.rundun.app. Each plan includes a storage allowance:

Storage add-ons are available at any tier (10GB/$0.50/mo, 50GB/$2.50/mo, 200GB/$10/mo, 1TB/$50/mo). Photo count never affects run cost — run credits and storage are billed separately.

At 100% storage quota, new runs are paused. Existing data is untouched and always exportable. See Pricing for full details.

Photo estimation

Based on 500KB per photo after device compression from typical 4MB camera output.