Shot sequence and Video Director

The Video Director and shot sequence tools make narrated motion-graphics videos: kinetic typography, shapes and images that appear the moment the voiceover says their phrase. The Video Director tools do it in one call; the shot sequence tools expose each step. This page lists the parameters. Video Director and Shot sequence explain how to use them.

start_video_director

Returns the motion-director guide an assistant follows to author a narrated video itself: how to pick a genre and a story arc, write the voiceover as cue phrases, build a shot sequence brief, and run the pipeline. It changes nothing and can be called any number of times.

This tool has no parameters.

create_explainer

Writes and renders a narrated, time-coded explainer video about a topic, in one call. The director writes the voiceover script and the shot sequence brief, generates the voiceover, aligns it word by word, and renders the MP4.

Permission: workflows:execute. Credits: the authoring run plus the voiceover, the alignment and the render; list_models lists the price under video-director. Nodaro Cloud only.

Returns: a job id. The card shows the progress and plays the video. Without cards, poll get_job; the MP4 is at output_data.videoUrl.

create_launch_video

Writes and renders a narrated product-launch video from a description of the product, in one call.

Permission: workflows:execute. Credits: as create_explainer. Nodaro Cloud only.

Returns: a job id.

The brand object

list_shot_shapes

Returns the catalog of the 15 shot sequence blueprints: ready-made animated beats such as kinetic-type-beats or titlecard-reveal, with their roles, description and default length in frames. Call it before you write a brief, to choose a blueprint for each beat.

This tool has no parameters.

get_shot_shape

Returns one blueprint in detail: its roles, description, default length, the JSON schema of its params, and a filled example you can paste into a brief.

list_brand_presets

Returns the 8 brand presets, such as midnight-violet and editorial-cream: a named palette and font pairing with its label, mood, description, palette and fonts. Pass a preset id as brand to keep every scene consistent.

This tool has no parameters.

forced_alignment

Aligns a known transcript to an audio or video clip and returns the start and end time of every word. A video's speech track is extracted first, so a talking-head clip can be aligned directly. It works like the Forced Alignment node.

Permission: workflows:execute. Credits: 30.

Returns: a job id. The alignment is in output_data.alignment: a list of { word, start, end }, in seconds.

resolve_shot_sequence

Bakes a shot sequence brief and the word timings into a render-ready plan: it matches each cue to its words and turns every cue-based reveal into an exact frame. It runs at once and returns the plan, with no job.

Permission: workflows:execute. Credits: free.

Returns: plan and warnings, for example a cue that was not found and fell back to proportional timing. When the scenes overlap in time, the tool returns 422 scene_overlap. A brand logo image is not added automatically here: include a logo-assemble-lockup reveal yourself to show it.

render_shot_sequence

Renders a resolved plan to an MP4. The video is also saved to your library.

Permission: workflows:execute. Credits: as the Render Video node: 50 up to 1920 pixels on the longest side.

Returns: a job id. The card shows the progress and plays the video.