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

# Search a Video

> Working with one specific video once you've found it.

There's no endpoint that scopes a search query to a single `video_id` — search always runs against a collection or your whole library, described in [Search a Collection](/search/search-a-collection). Once you have a `video_id`, either from a search result or from listing a collection, use it to fetch that video directly:

```bash theme={null}
curl https://api.pureframe.ai/v1/videos/vid_def456 \
  -H "Authorization: Bearer pf_..."
```

```json theme={null}
{
  "data": {
    "id": "vid_def456",
    "filename": "footage.mp4",
    "collection_id": "col_abc123",
    "status": "done",
    "duration_secs": 612.4,
    "playback_url": "https://...",
    "thumbnail_url": "https://...",
    "segments_count": 34
  }
}
```

`playback_url` is a presigned URL valid for about an hour — fetch a fresh one each time you need to play the video rather than caching it. See [Metadata](/video-processing/metadata) for every field on the video object.

## If you want "more like this, in this video"

If a search already returned segments from the video you care about, you have the `video_id` for each — filter the response client-side rather than making a second request. For finding visually similar moments across the same library (not scoped to one video), see [Image Search](/search/image): submit one of the video's own frame thumbnails as a reference image.

## Deleting a video

```bash theme={null}
curl -X DELETE https://api.pureframe.ai/v1/videos/vid_def456 \
  -H "Authorization: Bearer pf_..."
```

Permanently removes the video file, all frame data, and indexed segments. Returns `204 No Content`. This cannot be undone.