How to Make a Finished Video Without Any Editing Skills

You type what you want, attach whatever you already have, and a finished video comes back in the same conversation for you to watch before you download it. That's the whole shape of it, whether what you're handing over is a folder of clips off your phone, a document about your business, a link to something you wrote, or nothing at all except the idea in your head.

None of it opens an editing program. You never drag a clip onto a strip, never pick how long two shots should overlap, never hunt through a menu to find where the export button hides. You type the request the way you'd brief a person, and it gets to work while you go do something else: check back whenever you like, or close the tab and let it keep going without you.

Do you have to learn a video editing program first?

No. There's no editing screen to learn here in the first place, just a chat. Whatever the job, the pattern holds: say what you're after, and a finished file lands back in the thread. No separate program to open, no settings to guess at, because the file already comes out in the shape you need.

That doesn't mean any five words get you a good video. A message that just says "cut this down" gets you a cut. One that says "cut this to sixty seconds and drop the Q and A" gets you exactly that.

It also isn't a stack of separate programs you'd have to open one after another for cutting, then captioning, then a thumbnail. It's one conversation that can do all three, remembers what you already told it, and keeps working even if you close the tab and come back to it later. You can watch the actual work happen instead of guessing whether it's stuck.

What do you actually hand it?

A phone full of clips and no idea what's usable is the most common one. Attach the whole folder and ask it to go through everything: a saved instruction called Rough Cut From Footage watches every clip start to finish, sorts out what's worth keeping and what isn't, and lines up what's left in roughly the order the story should run. Send it one short clip on its own instead of a pile and it stops to ask you for the rest, since there's nothing to sort out of a single file.

Want short clips out of one long recording instead? A different instruction handles that: it watches the whole thing, picks two to four moments that stand on their own, and tells you in one sentence why it picked each one, rather than handing you a pile with no reasoning attached.

A PDF, a slide deck, a case study, a one-pager: any document hands over the same way. It reads every page rather than skimming the first few, pulls out the actual point being made along with two to four supporting facts, and builds a script and visuals around that point. If the document doesn't really have one, it says so instead of producing a good-looking video about nothing.

A link works much the same way. Paste the URL of an article you wrote, and it goes and reads the whole thing rather than guessing from the headline. If the link won't load, it asks you to paste the text in rather than inventing what the article probably says.

Nothing filmed and nothing written is still enough to start from. Say what you want explained or shown, and it researches the subject first, before any script gets written, then builds a short animated video from scratch around what it found. No clips required.

Whatever you're handing over goes in the same message box: attach the file, paste the link, or just type. Each of these routes starts life as a saved set of instructions you can read in full first.

What do you get back?

A finished video lands right in the thread, and you can open and play it there before you commit to downloading anything. Nothing on it carries a logo, ours or anyone else's.

What comes with it depends on the job. A cut built from raw footage arrives with a plain-text note listing which clips made it in, which ones got left out and why, and what still looks to be missing, so you've got something specific to argue with rather than a vague sense that it's off. Short clips come back with a sentence each on why that moment got picked. A video built from a document comes with the narration script as text, plus a one-line note on which part of the document it built around.

You say the shape in plain words, not a number. Ask for something built for a phone feed and it fits people and text to that upright frame; ask for something like a YouTube upload and it stays wide, no black bars added to force one shape into the other.

What if it's wrong?

Say so, in the same conversation, and it picks up the file already sitting there rather than sending you back to the start. "Trim the last clip, it runs long." "Go back to the second page of the document, you skipped the pricing." "Make it square instead of tall." None of those needs a fresh upload, and a trim like the first goes back through the same editing that made the cut.

Sometimes it stops and asks you first, rather than guessing on something you'd rather decide yourself: phone shape or wide, keep the introduction or cut straight to the point, three or four clips or just the two strongest. The footage-sorting instruction in particular won't start cutting anything until it knows what the video is for, about how long it should end up, and who's going to be watching it, unless your opening message already covered that.

One thing you can't do is interrupt a step already running. It finishes what it's doing, or you stop it and steer the next message instead.

What does it cost?

Signing up gets you five hundred credits on day one: credits are the unit that tracks how much you've used. After that, thirty more land each day, and the running daily balance stops at two hundred and ten. The five hundred you started with never expires.

Heavier or more frequent use has two paid plans above the free one: Starter at nineteen dollars a month for two thousand credits, and Pro at forty-nine dollars a month for six thousand. There's also a ten-dollar top-up for a thousand credits, for when a single project needs more than the daily drip covers and a subscription doesn't make sense yet. Opening an account takes nothing but an email address.

Where does this stop working?

Each video file needs to be under 200 MB to be watched and worked with directly. A phone clip usually clears that easily; a long, high-resolution recording off a proper camera might not. A single pass over a video also tops out around thirty minutes at the sampling rate used by default, so a longer recording gets read in stretches rather than all at once.

If a recording has more than one person talking, there's no automatic way to tell who said what. Nothing separates speakers on its own, so a two-person interview that needs each voice labeled correctly works better as two separate files, one per person.

Editing or animating a real person who's already in your footage is blocked outright for accounts in the UK, Switzerland, and the EEA. That's a permanent rule from the provider behind the tool, not a setting anyone here can turn back on; footage and images without people, or a video generated from a written description, aren't affected.

And once the file is finished, that's where it stops. Nothing gets posted to YouTube, Instagram, or anywhere else on your behalf. You get the download, and putting it up is on you.