Vidmoat is the video editor built for agents. Tell the built-in one what you want — or plug in Claude, Cursor, or your own — and it cuts, captions, grades and ships finished videos. You take the credit.
No credit card. No watermarks. No "rendering…" existential dread.
It finds the good takes, kills the dead air, and stitches the story — before your coffee even brews. Every cut stays editable, because we know you'll tweak it anyway.
See it in the editor →Never opened a video editor? Doesn't matter. Tell the agent "make this a punchy TikTok" and it reads your footage, writes the captions, cuts the dead air, and grades the color — step by step, out loud, until it's done.
The first video editor any AI agent can drive end-to-end. Connect Claude Code, Cursor, or your own agents over Model Context Protocol — they import footage, edit the timeline, look at the frames they made, and render finished MP4s. No browser. No clicks.
WORKS WITH EVERY AGENT THAT SPEAKS MCP
…and anything else with an MCP client. One key, one command to connect →
A full professional editor underneath the agent — open a chapter.
One click. No green screen. The background just… leaves.
Paint over anything you regret filming. Logos, strangers, exes.
Chroma key with spill cleanup. Pick the color, gone.
Every word lights up as it's spoken. Auto-synced to your audio.
Finds the good takes, kills the flops. Your rough cut, already cut.
Delete a sentence in the transcript, it disappears from the video. Editing by backspace.
Word-perfect, on-beat, styled to your brand. Not a court transcript.
Umms, ahhs, and dead air — detected and deleted. Ruthlessly.
One long video in, twelve vertical bangers out. Hooks included.
It scans your library and drops the right shot at the right beat. You take the nap.
Same agent, three completely different workflows. Yours is in here somewhere.
Records Tuesday night. By Wednesday morning: the full episode on YouTube, twelve clips for Shorts, karaoke captions on every one. She hasn’t opened a timeline in months.
Films the whole cook on one phone, dumps the footage, types “fast recipe edit, keep the flip.” The agent cuts the prep, keeps the fire, captions the seasoning rant.
Green-screens himself over client transformations, brushes out the gym’s logo, ships a polished breakdown to four platforms — before his next session walks in.
Drag in everything. The good takes, the bloopers, the 40 minutes you forgot to stop recording.
"Fast-paced cooking video, punchy captions, keep the part where the pan catches fire." Done.
Tweak what you want, export in every format, and get back to the part you actually love: filming.
For testing the waters. Or the moat.
For people who post like it’s their job. Because it is.
For teams shipping content at industrial scale.
No — that's the whole point. Type what you want in plain words (“make this a punchy TikTok”, “add captions”, “cut the boring parts”) and the AI agent plans and applies the edits. Everything stays on a normal timeline, so you can tweak by hand whenever you want.
Yes. 3 projects, the AI console, auto-captions, effects, and 720p exports — free forever, with a small watermark. Upgrade when you want 4K, no watermark, and the pro effects.
Your uploads live in your account and are only used to render your videos. The clip analysis (transcripts, silence detection) runs to power your edits — we don't train on your footage or share it.
Those are tools you operate. Vidmoat is agent-first: an editor that operates itself — an AI with hands on the actual timeline. It cuts, captions, grades and re-formats across platforms from one prompt, everything stays fully editable, and it's the only editor external AI agents can drive too.
Yes. Vidmoat ships an MCP (Model Context Protocol) server: create an API key in your dashboard, add it to Claude Code or Cursor in one command, and your agent can import media, edit the timeline with 65+ commands, preview frames as images, and render MP4s — without ever opening the app.
Anytime, in one click from your account page (Stripe handles it — no email-us-to-cancel games). Your plan stays active until the end of the period you paid for.
Auto-captions with word-sync, silence cutting, multi-step edit plans from one prompt, background removal, green screen, a cleanup brush, 33+ effects, smart splits with varied transitions — and the list grows weekly.
Takes 40 seconds. Less time than your last umm.
Also on desktop — edit 10 GB files with zero uploads