Guides

    Snapshot an EDL after the structure is the structure, not while tasting clips

    save_editor_project stores timeline state. Snapshotting a pile of maybes, or saving before picture lock, freezes a draft that will teach next week's edit the wrong footage.

    Versely Team3 min read

    Most recurring content is the same edit with new footage: a fixed intro, a fixed bed, captions in a fixed style, a middle that changes. save_editor_project stores that as cloud project state — clips, trims, texts, music, captions. Omit id to create; pass id to update. list_editor_projects returns metadata only (id, title, duration, clip count, thumbnail, updated_at). get_editor_project loads a timeline. edit_video is what makes it watchable. Saving is cheap. Rendering is the meter. Snapshotting a tasting bin is how next week opens on last week's misses.

    Save a video edit as a reusable draft is the job. Preview at 480p (preview:true, cooldown) before a charged export. The AI video editor is the surface.

    A draft is not a movie

    Nothing is watchable until edit_video. Teams save first because "we should version." Version of what? If the middle clip is still a maybe, the EDL is a mood. Next week's operator will load it, swap nothing, export, and ship last week's reject with this week's date.

    The create-versus-update rule is easy to get wrong. Omit id and you fork. That is either the copy you wanted or a duplicate you did not. Pass id to update in place. Learn that on a structure you intend to reuse, not in a panic save of a taste session.

    The EDL is Versely's project format, not an interchange file for another NLE. Do not treat save as an export.

    What to freeze

    Freeze the structure: intro slot, caption style, bed, duration math. Swap the middle on a keeper each week. Do not freeze specific miss URLs as if they were the template.

    1. Sign a first week's assembly (keepers).
    2. Save with a title you will recognize (Weekly product reel).
    3. Next week: listget → swap the middle for this week's keeper → save with id → preview → export.

    estimate_cost on the export, not on the save. Snapshotting before render is the cheap half — of a workflow whose expensive half should only see keepers.

    FAQ

    Can I inspect a timeline from the list?

    No. Metadata only. get_editor_project for that id. Listing is not a reason to save ten tasting versions. You will not open them.

    Should I save after every generate?

    No. Save when the shape is the shape. Generating is sampling. Saving is memory. Memory of misses is how templates rot.

    Does saving caption the video?

    No. Captions in the EDL render when you export. They should be the house style on a structure you believe, not a DYNAMIC experiment glued to a maybe.

    Why is this a later step if save is free-ish?

    Because the draft governs later exports the way a brand kit governs caption burns. Freeze the wrong structure and you industrialize the leak.