How Vibe Works

This page is the reference behind Meet Vibe.

Talking to it

Type your request in the composer and press Enter to send. Shift+Enter adds a newline. Press Escape once to arm a clear, and again to start a fresh chat. The composer shows Starting agent… while Klipeo establishes the durable run. Your text and attachments remain in place until the server accepts that run; if startup fails, correct the problem and press Send again without rebuilding the request.

Checkpoints are saved automatically by the harness; they are not tools you need to request. After reconnecting, matching editor receipts can be retried without applying the same edit again.

New editor sessions apply a tool's edit batch and confirm its composition save before reporting the tool as completed. If saving fails, keep the tab open and resolve the save error; a visible local edit is not yet a saved edit. If the editor cannot confirm it, Vibe stops further edits and asks you to reconnect and inspect the timeline; retrying blindly could duplicate changes.

Inserted, overwritten, and appended clips retain their IDs across follow-up edits. Insert shifts unlocked lanes together; overwrite clears only the target lane's covered span. Detaching audio returns a separate audio clip and mutes the video; it does not automatically group the two clips.

A few things you can do while writing:

Voice input uses the same composer on desktop and phone. It stays open while you speak even if a local agent connection changes; sending the finished request is still subject to the selected agent being available.

If opening the microphone takes more than 15 seconds, Vibe stops the attempt and shows a retry message. A take runs for up to three minutes and keeps going through pauses: if the transcription service drops the connection while you think, Vibe reconnects underneath and keeps what you already said. After you finish speaking, realtime transcription has an 8-second deadline; uploaded recordings have a 50-second deadline. You can cancel while waiting. Cancelled or timed-out takes never insert a late transcript into your draft.

Keep your asks concrete. "Trim the dead air and add TikTok-style captions" lands better than "make this good," because Vibe can map it to real operations. There's a longer guide to phrasing in Skills and cost habits in Save on AI cost.

It knows where you are

Each message carries what you are looking at: the page, the editor tab (Video, Script, or Moodboard), the panels you have open, the playhead, and the clips you have selected. So "make this one shorter" or "what's on this panel?" works without naming the clip or the tab. The crew reads this as a description of your screen, never as an instruction, and it only sees what is in Klipeo itself.

Briefs, brand, and questions

Vibe stops when it needs your own input, such as a choice only you can make, and carries on once you answer. For big jobs it writes a checklist and works through it; see The plan, and what it remembers.

For a new video or a substantial restyle, Vibe checks your brief, references, and saved style before editing. If an essential choice is missing, it asks a short question with concrete options. A clear edit proceeds directly. Local engines show these questions in the same composer as the cloud engine.

Naming a platform sets the frame. Ask for an Instagram, Reels, TikTok, or YouTube Shorts video and a new project switches to 9:16 before anything is built. Ask for a YouTube video and Vibe asks whether you mean a Short or a regular 16:9 video. If you name a shape yourself, such as "a square Instagram post", Vibe uses yours. A cut already on the timeline is reframed instead.

Vibe reads your owned Brand Design kits on each turn. Your default applies unless you ask for another style. Without a default, Vibe can offer up to six saved kits, including their private style guidance, when a new video needs a direction. Other kits remain available in Brand Design.

Reopening a chat restores its current unanswered question. Replaying saved history does not resubmit earlier approvals or start their edits again.

What happens after you hit send

Vibe works in the open. You'll see it think, run its tools, and change the timeline as it goes, rather than going quiet and dropping a finished result on you.

Because every change streams in as it happens, you can stop reading the summary and just watch the preview. If something looks wrong, you don't wait for the end. You fix it right there.

Chat text can continue arriving while an editor refresh is pending. Timeline changes still apply in order. If the connection drops, Vibe reconnects to the existing run rather than starting the job again. The transcript says Waiting for updates while live updates are unavailable, then returns to the reported activity when the connection is restored. If the transcript has not loaded yet, the chat shows a recovery message instead of the new-conversation suggestions.

A temporary error while recovering a saved run does not cancel that run. Vibe keeps its resume record and tries again when the browser comes back online. If recovery still fails, reload the editor after checking your connection.

Saved run events load from the database before the live event connection starts. A deployment can use a separate event gateway for the live connection. Your existing sign-in authorizes each reconnect; there is no second account to connect. Reloading or closing the stream does not send Stop to the agent. A slow live connection does not hold up that saved history. If live delivery loses an event or disconnects, Klipeo reads the missing committed events from the database before continuing. While replaying from the database, Klipeo retries an unavailable live transport after a five-second cooldown. You do not need to refresh to trigger that retry.

You can send a correction while Vibe works: press Enter, or the send button that appears beside Stop once you type. The draft stays in the composer until the server accepts it. Then the message appears in the chat, dimmed, with Queued · Vibe reads this at its next step. When Vibe reads it, the mark closes into a check with Read by Vibe, and Vibe's answer to it starts as a new reply below your message. Until then, Edit takes the message back into your draft, added after anything you have already typed. Text-only corrections with unchanged settings can enter at the next safe model step, after the current tool call finishes. Attachments, changed plan settings, and approval answers remain separate turns. Sending a correction does not stop the current job or bypass an approval. Queued corrections are restored when you reopen the active run. If a request times out, retry the unchanged draft to check the same submission. A newly selected preview target must wait for a separate turn because corrections use the active run's editor context.

While a reply is being written, its text, tool groups, generated media, and delivery cards appear in the order they happen. You can expand or collapse each tool group while the agent works. The current group stays open between calls unless you collapse it yourself; earlier groups collapse when the reply moves on unless you opened them manually. Once the reply finishes, its work folds into one Worked for 1m 5s · 12 steps summary that starts collapsed: the tool groups and the notes Vibe wrote along the way sit inside it, and the answer, generated media, and delivery cards stay in view below it. Open the summary to see every step in order. This also applies when you reopen a saved chat. Older replies retain their previous layout. Failed-step counts remain visible in the collapsed summary; unfinished steps are not marked as successful when a stream stops. Text and tool-start updates do not wait for editor refreshes. One shimmering status row remains visible throughout the active run, including between tool calls. Its label follows the actual phase: Catching up, Reading your timeline, Thinking it through, Writing the reply while the answer streams in, or what the current step is doing, such as Generating a video shot or Looking at the media. Every step in a tool group reads the same way, in plain words while it runs and in the past tense once it finished. It does not cycle through invented editing or rendering activity. Reduced-motion settings keep the label readable without shimmer.

Pending image, video, and audio work uses the same grid-reveal loading card. Each active media tool keeps its own card, including while another result has already arrived. Transcription, media import, export, and delivery steps use the same treatment while their tool is running; a completed or failed tool removes its pending card.

Text corrections can also be consumed at the next step of an approval-authorized run. A correction does not inherit the earlier approval: newly requested effects still pass the normal action-specific permission checks. A provider call already in progress finishes before the next step can consume your correction.

Long cloud conversations can create a model-generated context handover at a safe step boundary. Klipeo saves the handover before replacing the model context and retains active skill references and the files attached to the current request. A handover never happens before the model has seen your request. This does not remove the visible chat history.

Sharing a chat

Select Share in the agent header, choose invited people or anyone with the link, then select Create share link. Chats are private until you explicitly create a link. Invited readers must sign in with a verified email matching the invitation; sending email invitations is not automatic.

The read-only page shows user and assistant messages and tool status in pages of 200 messages. Select Older messages to continue through history. It includes future messages added to this chat, but does not grant editor access or expose private media and raw tool inputs or outputs. Public readers can copy the visible text. Revoke link disables future access; it cannot erase copies someone already saved. Creating a new link after revocation does not reactivate the old link.

Sharing requires its database migration to be deployed. If access cannot be loaded or a change cannot be confirmed, close and reopen the dialog to check the saved state before retrying. Chat-share pages and their bearer links are excluded from client analytics and replay.

Continuing an edit

If Vibe needs something from you mid-turn, it asks. The question shows up in the composer as a small wizard: a text box, a single-choice list, or a multi-select, depending on what it needs. When it asks several questions, Next moves to the following one and Continue on the last question sends your answers, so the turn only carries on once you have answered them all.

Anything Vibe generates during a turn (images, video, audio) shows up inline as a media card in the transcript, and collects in the media gallery so you can find it again later without scrolling back.

Vibe can also rework a clip that is already in your media bin: change its look or light, continue it past its last frame, or swap a person in it for the one in your photos. Each result is a new clip in the bin, so the original stays as it was.

Vibe looks at its own work. After a visual edit it captures stills from the preview and looks at them before saying the edit worked, and it looks at every picture or clip it generates before placing it, so a warped hand or a wrong product gets caught and redone. Ask what is in a clip or photo in your media bin and it looks at that too. Preview stills are saved privately with the chat so their cards load after a refresh; see Chats and history.

The Outputs card distinguishes downloadable files from render submissions. A render receipt confirms that the request was submitted, not that a finished video is ready. The card checks the render job and shows queued, rendering, uploading, completed, failed, or cancelled status. A completed job with an available download gets a Download video action. If checking fails, use Check again; this only refreshes the status and does not submit another render.

Several agents at once

A turn can run several tools in the same round, and each kind of work shows up as a named character. It goes to the real place the work is happening: the clip or track on the timeline, a card in the media panel, or the preview. It stays on that spot as you scroll or zoom, and shows what it is doing in its eyes. Work with no place on screen stays hidden rather than parked somewhere random. A character smiles and leaves when its job ends. One that fails winces and holds for a moment, and the error stays in the reply. Free edits that can be undone, such as cuts, trims and template placements, start as soon as Vibe writes them, so the timeline fills one step at a time. While any of them works, the editor header shows their faces and names, such as "Scout and Echo are working".

To call one yourself, type @ anywhere in the message and pick it from the Crew list, or open with its name, as in "hey Echo, add some music". That character takes the request. If you name more than one, each takes the part that fits. You don't have to name anyone: Vibe picks the right character for the work, and decides by itself when a job is worth spawning workers for.

A character can open the panel its work belongs to by pressing the real tab, such as Captions or Assets. It only ever presses those tabs. It never presses anything that edits, deletes, publishes, or spends, and it waits while you are using the editor.

The characters appear wherever the timeline does, so on a window under 1024 pixels wide you see them in the reply instead.

Generated shots don't wait for each other. When a request needs several clips or pictures, Vibe starts up to four at once and they render side by side, so a three-shot ad takes about as long as its slowest shot. In article mode, where each result also lands on the moodboard, generations still run one at a time.

For bigger jobs Vibe hands work to worker agents. When each shot needs its own make, check, and fix loop, it gives up to four workers one job each. They work at the same time, look at every result, redo what is wrong, and report back which clip or picture to use; Vibe then places the results on the timeline itself. Workers make new media and read your project but never edit the timeline. Each worker's steps show in the reply as their own group under the job's name, such as Shot 2 · waves · 4 steps, and its character carries the same job on its name tag from the moment Vibe hands out the jobs. Two workers can share a character. The groups stay the same when you reopen the chat. They generate under the same approval as the rest of the request. A correction you send while workers run reaches Vibe after each worker's current step: every worker stops there and reports what it finished and what is left, so you don't wait for the whole job. A correction that needs its own turn, such as one with an attachment or changed plan settings, lets the workers finish.

The plan, and what it remembers

This is the checklist Vibe keeps for itself during a run. For work with several meaningful stages, Vibe writes a checklist and ticks the steps off only when tool results confirm them. The latest checklist is stored with the chat, so it survives a reload without stacking stale earlier versions.

When you state a lasting preference, such as a house style or an audience, Vibe saves it with a remember step into your AI memory in Settings, and every future chat sees it. Say "forget that" and it removes the line. Inside a project, facts about the project go to the project's Memory tab instead.

Checkpoints and rewind

Before every cloud turn edits the composition, Vibe saves a checkpoint. The newest 20 per composition are kept, and each shows as a step in the transcript. Ask Vibe to rewind to one and it replaces the composition with that saved state as one undoable edit.

What the crew remembers

Press the brain icon at the top of the Vibe panel to see what the crew has saved: About you (lasting preferences, such as the platform you make videos for) and This project (decisions for the open project, such as the brand colour). Press the pencil to rewrite a fact and Save, or the bin to delete it. The crew reads both lists on every message, so a change applies from your next one. Notes you write yourself stay in Settings → Memory.

The first time you open Vibe, Scout says hello and offers three starters, each from the crew member who does it.

It opens things for you

Ask Scout to show you something and it opens it: "open the color page", "show me the effects", "take me to my renders". In the editor it can switch the tab (Video, Script, Moodboard), open a tool panel or an Inspect section, move the playhead, and select clips. It also opens Klipeo pages such as Projects, Crewmates, Templates, or Pricing; you go there when the turn ends. A message that only asks to go somewhere, such as "open the mixer", opens in about two seconds, because Klipeo recognises it and opens the place without waiting for the full agent.

It can also draw on the moodboard: "frame the hero shots", "draw an arrow from the logo to the colour swatch", "add a heading that says Launch look". It draws rectangles, ellipses, arrows, lines and text as ordinary board shapes, switches to the Moodboard tab, and shows them. Move, restyle or delete them like any shape, and Undo removes the whole drawing at once.

To show you something on the timeline, a crew member points at it. It flies to the clip or moment, selects it, moves the playhead there, and leaves a short note beside its name, such as "This cut lands half a beat late". The note stays long enough to read, then the crew member leaves.

Ask it to change an editor aid and it does: turn snapping or ripple editing on or off, show the media panel as a grid or a list, or zoom the timeline to fit, in or out. Each is a view setting you can flip back with one click.

Settings and payment pages, such as Billing, Account, API keys, or Model keys, always ask first with an Open Billing for you? card, even on the cloud engine's full access. Choose Yes, open Billing to go there when the turn ends, or Not now to stay. Opening a page never changes an account setting; you change it yourself once you are there.

Style and sound in one turn

Before a turn starts, Klipeo measures what kind of video this is (a launch film, a screen demo, a podcast clip, and so on) from your timeline and your words, and the crew follows that style's rules. When the turn finishes editing, Klipeo checks the result against the style, such as shot length and caption readability, and the crew fixes only what failed. Both happen without slowing the reply.

You can build motion graphics and ask for their sound in the same message: the crew sound-designs titles and moves you just added, before you save. Sound placed from unsaved keyframe moves is timed from their keyframes, so it is checked again once the edit is saved.

Open it, and what to do next

When a turn makes or changes something worth seeing, the reply ends with a button that takes you to it, such as Inspect text or View renders, and up to two next steps the crew suggests. The button moves the playhead, selects the clip, and opens the panel or page that shows the work. A suggestion goes into the composer, so you can send it as is or change it first. Only the newest reply shows these; a reply that only answers a question has none.

Approvals

When a tool would change the timeline, publish somewhere, or spend credits, Vibe raises an approval that says what the tool will do. On the Klipeo cloud engine, every run has full access, so Klipeo answers these approvals for you and the run carries on. A request to click and type in the Klipeo window on a Mac always waits for you, and so does opening a settings or payment page and a question that needs your own answer.

Generation is approved once per request: the first image, clip, voiceover, music, or sound effect covers every one the request needs. The approval never covers replacing or deleting your own media, and a new message starts a new request. If you close the editor while a run is still working, reopening it reattaches to the run and any edit that was in flight lands.

On a local engine, Klipeo asks you to confirm the engine's access for the project before the first send, and publishing always asks first. Confirmation is per project and per engine; changing either asks again. See Engines and models.

Chats and history

Sessions are saved on the server, so a chat survives a reload. The Vibe header has a recent-chats dropdown where you can load an old chat, rename it inline, or delete it, and a New button to start a clean one.

Preview frames Vibe captures are saved privately with stable, signed-in image links, so their chat cards load after a refresh. Older captures whose image data was not kept ask you to inspect the preview again. Shared chat links do not grant access to these private images. Capture waits for decoded video frames, and a failed capture is reported instead of being treated as proof that a visual change worked.

Finishing moves

Vibe drives the same editing operations you have by hand, grouped into areas like captions, cutting and trimming, audio cleanup, color, transitions, zooms, voiceover, and full end-to-end edits. The curated recipes for the common jobs are listed in Skills. If you can describe an edit in terms of those, Vibe can usually do it.

Talking footage gets the same Smart Cut as the editor's panel. Ask Vibe to tighten a clip, cut the ums and long pauses, or keep the best take of each line, and it picks the passes that fit, cuts what it is sure of in one undoable edit, and tells you what it left in.

A few finishing moves that editors usually do by hand take one sentence:

Ask for punch-ins before the vertical reframe: the reframe keeps each clip's zoom and follows the speaker at it. Cleaning up captions rebuilds them, so a style you set on one caption by hand goes back to the caption style.

Recordings made in Klipeo carry their cursor data to Vibe, so it can zoom where you clicked and cut the dead waits from what you actually did rather than guessing from the picture.

When something goes wrong

If a step fails, Vibe says what needs attention instead of pretending it worked. Progress describes the project task without showing internal tool names or raw technical errors. You can provide a missing connection, choose another source, or do that piece by hand and let Vibe carry on with the rest.

When an interrupted run resumes, Vibe reuses confirmed pictures, clips, and edits already saved in your project. An uncertain result still needs checking before the action can be repeated.

When an action has invalid inputs or needs another available tool, Vibe can correct the request and continue within a bounded recovery budget. Temporary failures on free reads can be retried once. An unconfirmed generation, edit, send, or publish needs its saved result checked before it is repeated.

Stopping prevents new work from starting. It does not undo an action that already reached a connected app. If delivery is uncertain, Vibe reports that uncertainty instead of claiming success or sending the same request again.

Where Vibe runs

By default Vibe runs on Klipeo's cloud, so there's nothing to install. Every cloud turn runs on Gemini 4 Argon with medium thinking effort, so there is no model to pick. You can also point Vibe at a coding CLI on your own machine (Claude Code, Codex, Gemini CLI, or OpenCode), which is free to run and lets you choose that CLI's model and reasoning level. The agent chip under the message box shows which one the next turn uses and switches between them. That's all in