Back Office · office.temerarii.xyz
One asset, all the way in — composition, the wireframe + storyboard, the output format stack, and the template, all read from the SAME content-index record. The expected output matches what /media surfaces for this post.
post social-W36-Thu-3kind threadweek W36date 2026-09-10campaign thread · Thupillar performancebeat Thuasset videoduration 48.5sground redscenes 9

Checklist the per-video bar — engine/sim

86.6/100
plain languagevo coverageno dead airuniquenesscaption fitcompletenesscleanliness
quantitative quality · weights learn from your reviews (engine.sim.memory review social-W36-Thu-3 good|bad)
⚠ 3 flag(s) — not yet ship-ready: copy_genericdup_sequence_across_assetsdup_set_across_assets · see docs/strategy/VIDEO-CHECKLIST.md

Composition comp · template family · expected output

composition SceneReelfamily / template SignalFieldReel
9:16 Reelpending1:1 Squarepending16:9 Widepending9:16 4Kpending1:1 4Kpending16:9 4KpendingGIF (SMS)pending
render pending — silent master not yet on disk
expected output: 0/7 rendered — same matrix the /media preview surfaces for this asset.

Composition layer × scene 9 scenes · 48.5s · comp_id + rendered still + tier + the script

#Layer (comp_id · still · tier)BeatTimecodeMotionLogoAudioVO / on-screen / caption
1s1
matches intent
shared field
signature-3d		open0–5.0sspatial-parallaxicon·white-knockout♪ bed_in
Marketplace listings used to just sit there silent, hoping to be found.
on-screen: Listings used to sit there silent
expected on screen: red ground · dodeca hero in the shared Signal Field · Mensor leads · node-graph · spatial-parallax · icon·white-knockout logo · caption bottom-left
spec (the prompt): comp_id shared Signal Fieldvisual node-graphshape dodecaground redtreatment white-knockoutmotion spatial-parallaxpower summoninstrument summon→Listings used to sit there silent
2s2
matches intent
shared field
signature-3d		hook5.0–10.2skinetic-buildicon·white-knockout♪ node_lock
You wrote one title, posted one photo, and never tested a thing again.
on-screen: One title, one shot, no testing
expected on screen: red ground · dodeca hero in the shared Signal Field · Mensor leads · mark · kinetic-build · icon·white-knockout logo · caption bottom-left
spec (the prompt): comp_id shared Signal Fieldvisual markshape dodecaground redtreatment white-knockoutmotion kinetic-buildpower laser-lockinstrument laser-trace→One title, one shot, no testing
3s3
matches intent
NumberedList
template		teach10.2–15.399999999999999skinetic-buildicon·liquid-chrome♪ node_lock
First move: build the title around the words buyers actually type to search.
on-screen: Write the title around real searches
expected on screen: red ground · a NumberedList panel over a dimmed Signal Field · Mensor leads · node-graph · kinetic-build · icon·liquid-chrome logo · caption bottom-left
spec (the prompt): comp_id NumberedListvisual node-graphshape dodecaground redtreatment liquid-chromemotion kinetic-buildpower morphinstrument morph+laser→Write the title around real searchescurate items, nodes
4s4
matches intent
ComparisonTable
template		proof15.4–20.6sreceipts-counticon·white-knockout♪ node_lock
Then have a model draft ten title variants from those exact search terms.
on-screen: Let AI draft ten title variants
expected on screen: red ground · a ComparisonTable panel over a dimmed Signal Field · Mensor leads · receipts · receipts-count · icon·white-knockout logo · caption bottom-left
spec (the prompt): comp_id ComparisonTablevisual receiptsshape dodecaground redtreatment white-knockoutmotion receipts-countpower receiptsinstrument spotlight→Let AI draft ten title variantscurate colA, colB, rows, statsLabels
5s5
matches intent
BuildLog
template		diff20.6–26.5scrossfade-8ficon·liquid-chrome♪ node_lock
But a title that is pure keyword soup reads like a robot and repels buyers.
on-screen: Keyword soup alone repels buyers
expected on screen: red ground · a BuildLog panel over a dimmed Signal Field · Mensor leads · code · crossfade-8f · icon·liquid-chrome logo · caption bottom-left
spec (the prompt): comp_id BuildLogvisual codeshape dodecaground redtreatment liquid-chromemotion crossfade-8fpower morphinstrument morph→Keyword soup alone repels buyerscurate codeLines, lines
6s6
matches intent
ChecklistCard
template		teach26.5–32.0skinetic-buildicon·white-knockout♪ node_lock
So you keep one clear human benefit right up front, then the keywords follow.
on-screen: Keep one human benefit up front
expected on screen: red ground · a ChecklistCard panel over a dimmed Signal Field · Mensor leads · node-graph · kinetic-build · icon·white-knockout logo · caption bottom-left
spec (the prompt): comp_id ChecklistCardvisual node-graphshape dodecaground redtreatment white-knockoutmotion kinetic-buildpower morphinstrument morph+laser→Keep one human benefit up frontcurate items, nodes
7s7
matches intent
KpiGrid
template		proof32.0–37.2sreceipts-counticon·liquid-chrome♪ node_lock
Now the listing gets found and clicked, not just found and scrolled past.
on-screen: Found and clicked, not just found
expected on screen: red ground · a KpiGrid panel over a dimmed Signal Field · Mensor leads · receipts · receipts-count · icon·liquid-chrome logo · caption bottom-left
spec (the prompt): comp_id KpiGridvisual receiptsshape dodecaground redtreatment liquid-chromemotion receipts-countpower receiptsinstrument spotlight→Found and clicked, not just foundcurate kpis, statsLabels
8s8
matches intent
WireframeMock
template		step37.2–43.5skinetic-buildicon·white-knockout♪ node_lock
Try it: pull real search terms and ask for ten titles, then pick the human one.
on-screen: Mine the search terms, draft ten titles
expected on screen: red ground · a WireframeMock panel over a dimmed Signal Field · Mensor leads · pipeline · kinetic-build · icon·white-knockout logo · caption bottom-left
spec (the prompt): comp_id WireframeMockvisual pipelineshape dodecaground redtreatment white-knockoutmotion kinetic-buildpower throwinstrument laser-trace→Mine the search terms, draft ten titlescurate stages, steps
9s9
matches intent
shared field
signature-3d		resolve43.5–48.5scoalescenceicon·white-knockout♪ bed_out
Soon listings rewrite themselves toward whatever buyers are searching this week.
on-screen: Soon listings rewrite for demand
expected on screen: red ground · dodeca hero in the shared Signal Field · Mensor leads · coalescence · coalescence · icon·white-knockout logo · caption bottom-left
spec (the prompt): comp_id shared Signal Fieldvisual coalescenceshape dodecaground redtreatment white-knockoutmotion coalescencepower coalescenceinstrument coalescence→Soon listings rewrite for demand

Format stack 3 aspects · same scenes[], re-cropped

9:16
1080×1920
Stories · TikTok · YouTube Shorts · Reels
1:1
1080×1080
LinkedIn · Facebook · Instagram
16:9
1920×1080
X/Twitter · YouTube · LinkedIn video

Channels 9 destinations

LinkedInX/TwitterYouTubeInstagramFacebookThreadsTikTokPinterestBluesky

Social captions supplemental published copy · per channel (comp_id level)

tiktokYour marketplace listing is invisible. Here is the move (AI-assisted): build the title around words buyers actually type, have a model draft ten variants, but keep ONE human benefit up front so it doesn't read like keyword soup. Found AND clicked. #marketplace #ecommerce #aitools
instagramYour listing is invisible. Build the title on real searches. AI drafts ten variants. Keep one human benefit up front. Found and clicked, not scrolled past. #marketplace #ecommerce #onlineselling #aitools #seo
linkedinMarketplace listings used to sit silent with one untested title. The method: build the title around the words buyers actually type, have a model draft ten variants from those terms, then keep one clear human benefit up front. The takeaway: pure keyword soup repels buyers, so a listing should get found and clicked, not just found and scrolled past.
xMarketplace move: build the title on real search terms, let AI draft ten variants, keep one human benefit up front. Found AND clicked, not scrolled past. temerarii.com
facebookMarketplace listings used to sit silent with one untested title. The method: build the title on words buyers actually type, have a model draft ten variants, and keep one human benefit up front. Found and clicked, not scrolled past. See it at temerarii.com.
threadsYour marketplace listing is invisible. Build the title on real search terms, let AI draft ten variants, keep one human benefit up front. Found AND clicked, not scrolled past.
pinterestAI marketplace listing optimization: how to get found and clicked. Build the title around real buyer search terms, draft ten variants with a model, and keep one human benefit up front. Evergreen guide to ecommerce, marketplace SEO, and product titles.
blueskyMarketplace move: build the title on real search terms, let AI draft ten variants, keep one human benefit up front. Found AND clicked, not scrolled past. temerarii.com
youtubeAI Marketplace Listings: Get Found AND Clicked Marketplace listings used to sit silent with one untested title and one photo. The real method: build the title around the words buyers actually type to search, then have a model draft ten title variants from those exact terms. Pure keyword soup reads like a robot and repels buyers, so keep one clear human benefit up front and let the keywords follow. Found and clicked, not scrolled past. More at temerarii.com.

Cross-links every lens is a view on this one record

→ Wireframe · scenes→ Calendar week→ Media (audience preview)→ Storyboard hub→ Day view→ Channel view