Assemble video
assemble_videoJoin clips the account already has into ONE finished video, in the order given, optionally over a voiceover (narrationUrl) and a music bed, which become separate tracks with the clips ducked under the voice, and with burned-in captions. The bed is a track from the library (musicUrl) or one composed for this cut from a description (musicPrompt), never both. One clip plus a track is how to put music or narration under a single video: "add music to my video" is clipUrls [that video] with a musicPrompt, and no pipeline to build. Every url must come from list_assets or a run's outputs in get_run. A slideshow of photos or images, with a push-in, a pan or Ken Burns, is ONE call with the image urls, stillMotion and stillSeconds, never an image-to-video clip per photo. Stills hold 3 seconds each (or stillSeconds), or stretch so a longer voiceover plays in full. With no stills, the clips have to run as long as the voice (list_assets gives each durationSec): the last frame holds up to 3 seconds so the voice can finish, and a voice longer than that past the clips is refused with nothing charged. lengthSec in the result is how long the cut will run, and any warning in it should be passed on. Renders in the background: poll get_run with the returned pipelineId and runId. Free once the account has bought credits (the result carries charged: false), so never ask a paying account to top up before assembling; charged on a trial. The same request made again within a few minutes returns the render already started (repeat: true) rather than starting another; to render the same cut a second time on purpose, pass a new requestId. Needs the pipelines:run and pipelines:write scopes.
Tool call template
Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.
{
"id": 1,
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"arguments": {
"clipUrls": "<clipUrls>"
},
"name": "assemble_video"
}
}Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| name | No | What to call the render, e.g. "Beach trip cut". | |
| captions | No | Transcribe the finished edit and burn captions in. Only worth it when the clips or narration contain speech. | |
| clipUrls | Yes | Video or image urls in play order: two or more, or exactly one when narrationUrl, musicUrl or musicPrompt is set. | |
| musicUrl | No | Music bed under the whole edit, beneath the clips' own sound and any narration: an audio url from list_assets. Leave it out when passing musicPrompt. | |
| fadeInSec | No | Fade the opening up from black over this many seconds: the first shot's picture and the music bed together. | |
| requestId | No | Any id you choose for this one request. A call with an id that already started a render returns that render; a new id renders again even when everything else is identical. | |
| fadeOutSec | No | Fade the whole ending to black and silence over this many seconds: the last shot's picture eases down and the music bed fades with it. This is "fade out at the end"; 1.5 to 3 reads as a deliberate ending. The transition argument is the cut between shots and does not touch the ending. | |
| transition | No | How each shot hands off to the next: the join between two shots, applied at every cut. Defaults to a hard cut. "fadeblack" dips to black between shots; it is not a fade-out at the end, which is fadeOutSec. | |
| callbackUrl | No | An https URL Treza POSTs the finished run to, once, when it ends (success, partial, error or cancelled), so an automation can wait for the render instead of polling get_run. The body is the JSON get_run returns for the run, sent with the headers X-Treza-Event: run.finished and X-Treza-Delivery: <runId>. It must be https on the default port and reach a public address; a URL that does not is refused before anything runs or is charged. Each attempt waits 10 seconds for a 2xx, and a timeout, 408, 429 or 5xx is retried twice more. | |
| musicPrompt | No | Compose the music bed for this cut instead of taking one from the library: genre, mood, tempo and instruments, e.g. "calm cinematic ambient, soft piano, gentle pads, no vocals". It sits and ducks exactly like a musicUrl bed and takes the same fades. Written by the Audio Generation node's default model (get_node_type music-gen), or its full-song model when the cut runs past 30 seconds; the result names the model as composedMusic. Leave it out when passing musicUrl. | |
| orientation | No | "vertical" is 1080x1920 for Shorts, Reels and TikTok, "horizontal" is 1920x1080, "match_first" (the default) keeps the first clip's shape. Shots that do not fill the frame are letterboxed, never cropped. | |
| stillMotion | No | A slow camera move over every still image: push_in, pull_out, a drift across the photo (pan_left, pan_right, pan_up, pan_down), or ken_burns (a different move on each photo in turn). Video clips play as shot. Defaults to none. | |
| narrationUrl | No | Voiceover laid over the whole edit. Everything else ducks under it. | |
| stillSeconds | No | How long each still image holds, in seconds. Defaults to 3; 3 to 5 reads well for a slideshow. A narrationUrl longer than the stills still plays in full: they stretch to cover it. Leave it out when the voice should set the pace of a cut that is all stills. | |
| audioFadeInSec | No | Fade the music bed (or the only track) up over this many seconds at the start. Applies to a composed bed too. | |
| audioFadeOutSec | No | Fade the music bed (or the only track) down over this many seconds at the end, with the picture left as it is; 2 to 4 reads as a deliberate ending. Narration is never faded when there is a bed. For the whole ending to fade, use fadeOutSec. |