Two hours for twenty seconds
The otter surfaced once, near the end of a two-hour locked-off take. Finding it means scrubbing through water and reeds at double speed, hoping you don't blink.
Locked-off takes from a blind, long-lens tracking at first light, a season of cards. Ask for every clip where the otter surfaces and get the in point, not the hour of empty water before it.
14 days free, then $59 a year or $7.99 a month. Works with your Claude or ChatGPT plan; timelines need DaVinci Resolve Studio 21.1+. Clip descriptions are optional and use your own API key, billed by Anthropic or OpenAI.
The otter surfaced once, near the end of a two-hour locked-off take. Finding it means scrubbing through water and reeds at double speed, hoping you don't blink.
Which card had the fox, which morning had the fog. The notebook says “about 6:40”, the camera clock says something else, and nobody wrote down the lens.
A wildlife film is shot over months and many locations. By the edit nobody can watch it all again, so good sightings stay buried on the drives.
Prewatch indexes every clip on your Mac, so Claude or ChatGPT knows what is in your footage and can build the cut with you in DaVinci Resolve Studio.
| What you get | Where it runs | What it does for you |
|---|---|---|
| Sightings, entrances and exits | Claude or OpenAI | Moments when an animal is seen, comes into the frame or leaves it, merged into usable ranges with in and out points. With your own Claude or OpenAI API key. |
| Honest about distant subjects | Claude or OpenAI | Species named only when clearly identifiable. A dark speck on the water is described as one and marked uncertain with a note, not guessed. |
| Birdsong and other sounds | On your Mac | Birds, insects, other animals, water and wind heard in each clip, tagged on the Mac by Apple's own sound classifier. |
| Empty takes flagged, quiet ones kept | On your Mac | Static, soft, dark and frozen takes flagged. A small animal barely changes a wide frame, so once a sighting is described, its take is not called static. |
| Days, places and camera facts | On your Mac | Clips grouped by capture time and location, with camera, lens, exposure settings and log profile read from each file where the camera wrote them. |
| Sequences built in Resolve | Your AI app | Claude or ChatGPT lays the chosen sightings on a timeline in DaVinci Resolve Studio 21.1 or later, through Resolve's own MCP server. |
You ask in Claude or ChatGPT. Prewatch answers with the clips, the seconds and the timecode, and your assistant builds the timeline in DaVinci Resolve Studio.
Each entrance across the season, with its in point, how long it stays usable, a still to check and a 1–5 usability score.
Your AI appThe clips where birds were heard, grouped by day and scene, so you can go straight to those mornings.
Your AI appA timeline in DaVinci Resolve Studio with the highest-scoring sightings, sharp and steady ones first.
Your AI appOnly when the description model can clearly identify it. Otherwise it describes what is visible and marks it uncertain with a note, such as “could be driftwood or a small bird”.
On the default model a typical clip costs a fraction of a cent and a long take about one cent, because only still keyframes, one contact sheet and the transcript text are sent. Settings shows the cost per clip before you turn it on.
Yes. Locations stay on your Mac as coordinates, with no lookup over the network, and are never sent to the description model. You can also switch off recording them in Settings.
Building timelines by chat needs DaVinci Resolve Studio 21.1 or later, whose built-in MCP server lets your assistant work in Resolve. With the free version you can still index your footage, search it and ask Claude or ChatGPT about it, then place the clips yourself. A file handoff that the free version can import is planned.
Point Prewatch at a folder you have already shot, then ask about it in Claude or ChatGPT.
14 days free, then $59 a year or $7.99 a month. Works with your Claude or ChatGPT plan; timelines need DaVinci Resolve Studio 21.1+. Clip descriptions are optional and use your own API key, billed by Anthropic or OpenAI.