varg-previz
Plan and pre-visualise any multi-shot AI video before spending credits on varg - music videos, ads and product promos, UGC-style spots, trailers, short films, explainers. Write the story (Hero's Journey, hook-problem-payoff, 3-act), lock character/product/vehicle sheets and one master plate per location, group shots into master scenes, block every camera in a three.js grey-box previz on the timeline (auto cameras from plain storyboard words like "MCU, 3/4 left, push in"), then export clean previz clips as @video1 references for Seedance 2.5 video-to-video and run one short generation per scene. Any aspect ratio, with or without a song. Use when the user asks for a previz, a 3D animatic, camera blocking or approval, shot lists, scene grouping, character or product sheets, story structure for a clip or ad, or a Seedance video-to-video pack.
varg previz: story → anchors → 3D previz → Seedance v2v
AI video models redraw the world on every call. A previz fixes what they are worst at, camera, geography, scale and timing, for free before the paid step. It is a cheap grey-box: it approves cameras and blocking, never the look.
Built on a 36-shot, 99 s music video ("I Don't Wanna Pick", 9:16, 133.8 BPM) and tested on a 6-shot product ad.
The app is in templates/previz/; scripts/new_previz.py sets it up in a project.
Pick the depth first
| project | do |
|---|---|
| 1–3 shots, simple camera | skip the previz: anchors (step 3) + image-to-video |
| 4–15 shots, one or two places, ad / UGC / promo | steps 1–3, auto cameras from the storyboard, a quick approval pass in the previz, pack |
| music video, trailer, short film, anything with geography, scale or many places | everything below |
The pipeline (in this order)
- Story: pick a structure for the format and map it onto the timeline (song sections, VO beats or seconds).
One story beat per section, the hook in the first 2–3 s. →
references/story-structure.md - Storyboard:
storyboard.jsonis the single source of truth for timing. Every shot getsid, start, end, scene, set, picture, action, camera, plusline(lyric/VO),beat,overlay,sourceas needed. Cuts on downbeats when there is a song. → schema inreferences/threejs-previz.md - Anchors: character sheet, product/vehicle sheet with a human for scale, antagonist sheet if any, ONE master
plate per location. Close-ups are generated from the master. Once the previz exists, make plates by image-editing
a previz render ("keep camera and geometry exactly") instead of from text. →
references/character-sheets.md - Master scenes: group shots by location; each scene becomes one or more 4–11 s Seedance runs.
→
references/scene-grouping.md - 3D previz:
python3 <skill>/scripts/new_previz.py <project>thenpython3 <project>/previz/serve.py. Build one grey-box set per location, check every auto camera, fix the bad ones in the UI, approve each shot, save, export the camera plan. Check frames in headless Chrome. →references/threejs-previz.md - Seedance pack:
build_pack.py --check, capture clean clips, build the pack (one @video1 per run + only the stills of what is in frame + prompts), review the board, estimate, then generate. →references/seedance-v2v.md
Spend nothing before step 6 except cheap anchor stills. Steps 1, 2, 4 and 5 are free, and the director's notes cost least there.
Hard rules learned the expensive way
- One location = one master plate = one set. Wide and close coverage of a place share the same set and refs, or faces, layouts and lighting drift between shots.
- References only for what is in frame. A sheet that also shows the monster pulls the monster into a shot that
should not have it, so crop. Never use an exterior still as a ref for an interior scene. When a shot is rewritten,
its old still is stale: flag it (
needs_new_still), do not feed it to the model. - Narrow frames hide things. Horizontal half-FOV = atan(tan(vFOV/2)·W/H). At 9:16 and vFOV 55° a subject 17° off-axis is out of frame. Check every camera in the map view: cameras end up inside walls, behind headrests, or inside a subject that walks into them.
- Scale is explicit. Put a human on every vehicle/product/creature sheet and state the ratio in every prompt (car ≈ 2.5× a person, ship ≈ 5×, monster ≈ hundreds of ships); otherwise models draw a man as big as his ship.
- No text from the model. "No text, letters, numbers, logos or UI" in every prompt. Lyrics, captions, prices, HUD and logos are drawn later in code (HyperFrames). Text labels are hidden in exported reference clips.
- Price before you promise. Run
estimate_coston one real request, quote it as the ceiling, report the billed cost after (get_job_cost). Checkcheck_balanceand which team the connection bills (get_current_team). - Ask before spending and before switching the billing team. Uploading and generating are outward actions.
Test one short run before a batch (
build_pack.py … --runs <key>). - Record the director's words verbatim in
brief.mdwith the date; apply each note to the storyboard, scene table, treatment and previz in the same pass so they never disagree.
Deliverables
brief.md(the director's words, dated),treatment.md(logline, structure with timings, colour script, recurring markers, references)storyboard.json+scenes.md(scene table: shots, cut time, gen length, refs, cost)previz/(three.js app,config.json,data/edits.json,camera_plan.md)previz/seedance/(pack.json,video/video_XX__<run>__<shots>.mp4,refs/,PROMPTS.md,requests.json,index.htmlboard,jobs.jsononce submitted,out/results)
Related skills
varg-ai (single generations, rendering), hyperframes (text, captions and graphics drawn in code on top of the
clips, final assembly), varg-music-video if installed (song measurement, lip-sync pass, platform masters).
- Downloads
- 71All time
- Security audit
- Review
- Last updated
- 3d ago
- Current version
- v1.0.0
- License
- MIT-0
