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-W33-Tue-3kind threadweek W33date 2026-08-18campaign thread · Tuepillar multimediabeat Tueasset videoduration 64.6sground whitescenes 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-W33-Tue-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 · 64.6s · comp_id + rendered still + tier + the script

#Layer (comp_id · still · tier)BeatTimecodeMotionLogoAudioVO / on-screen / caption
1s1
matches intent
shared field
signature-3d		open0–8.1sspatial-parallaxicon·color♪ bed_in
A model can now prep a whole interview for you, which frees you to do the one thing it cannot: listen.
on-screen: AI preps the interview, you listen
expected on screen: white ground · octa hero in the shared Signal Field · Lumen leads · node-graph · spatial-parallax · icon·color logo · caption bottom-left
spec (the prompt): comp_id shared Signal Fieldvisual node-graphshape octaground whitetreatment colormotion spatial-parallaxpower summoninstrument summon→AI preps the interview, you listen
2s2
matches intent
shared field
signature-3d		hook8.1–15.399999999999999skinetic-buildicon·wireframe♪ node_lock
Old way, you skimmed a bio in the hallway and winged the questions, then hoped for a good moment.
on-screen: The old way: wing the questions
expected on screen: white ground · octa hero in the shared Signal Field · Lumen leads · mark · kinetic-build · icon·wireframe logo · caption bottom-left
spec (the prompt): comp_id shared Signal Fieldvisual markshape octaground whitetreatment wireframemotion kinetic-buildpower laser-lockinstrument laser-trace→The old way: wing the questions
3s3
matches intent
NumberedList
template		teach15.4–23.5skinetic-buildicon·wireframe♪ node_lock
Here is the move: feed the model their past talks and writing, and ask it for the questions nobody else asks.
on-screen: Feed it their work, get sharp questions
expected on screen: white ground · a NumberedList panel over a dimmed Signal Field · Lumen leads · node-graph · kinetic-build · icon·wireframe logo · caption bottom-left
spec (the prompt): comp_id NumberedListvisual node-graphshape octaground whitetreatment wireframemotion kinetic-buildpower morphinstrument morph+laser→Feed it their work, get sharp questionscurate items, nodes
4s4
matches intent
RankList
template		proof23.5–29.4sreceipts-counticon·wireframe♪ node_lock
Sharper questions pull realer answers, because the guest can tell you actually did the homework.
on-screen: Better questions, realer answers
expected on screen: white ground · a RankList panel over a dimmed Signal Field · Lumen leads · receipts · receipts-count · icon·wireframe logo · caption bottom-left
spec (the prompt): comp_id RankListvisual receiptsshape octaground whitetreatment wireframemotion receipts-countpower receiptsinstrument spotlight→Better questions, realer answerscurate rows, statsLabels
5s5
matches intent
JsonDiff
template		diff29.4–36.0scrossfade-8ficon·wireframe♪ node_lock
A model can write the questions. Hearing the surprising thing they just said is still all you.
on-screen: Listening is the rare skill
expected on screen: white ground · a JsonDiff panel over a dimmed Signal Field · Lumen leads · code · crossfade-8f · icon·wireframe logo · caption bottom-left
spec (the prompt): comp_id JsonDiffvisual codeshape octaground whitetreatment wireframemotion crossfade-8fpower morphinstrument morph→Listening is the rare skillcurate codeLines, fileName, lines
6s6
matches intent
ChecklistCard
template		teach36.0–42.6skinetic-buildicon·wireframe♪ node_lock
Now run the recording through a transcript and have the model mark the three most quotable moments.
on-screen: Pull the clips by transcript
expected on screen: white ground · a ChecklistCard panel over a dimmed Signal Field · Lumen leads · node-graph · kinetic-build · icon·wireframe logo · caption bottom-left
spec (the prompt): comp_id ChecklistCardvisual node-graphshape octaground whitetreatment wireframemotion kinetic-buildpower morphinstrument morph+laser→Pull the clips by transcriptcurate items, nodes
7s7
matches intent
StatScoreboard
template		proof42.6–49.9sreceipts-counticon·wireframe♪ node_lock
It hands you timestamps for the gold, and your edit drops from a whole day to a single afternoon.
on-screen: Mark the gold, cut the clips fast
expected on screen: white ground · a StatScoreboard panel over a dimmed Signal Field · Lumen leads · receipts · receipts-count · icon·wireframe logo · caption bottom-left
spec (the prompt): comp_id StatScoreboardvisual receiptsshape octaground whitetreatment wireframemotion receipts-countpower receiptsinstrument spotlight→Mark the gold, cut the clips fastcurate pillar, stats, statsLabels
8s8
matches intent
PipelineMap
template		step49.9–56.9skinetic-buildicon·wireframe♪ node_lock
Three steps: let the model prep sharp questions, you listen for the surprise, then it marks the clips.
on-screen: Step: prep, listen, clip
expected on screen: white ground · a PipelineMap panel over a dimmed Signal Field · Lumen leads · pipeline · kinetic-build · icon·wireframe logo · caption bottom-left
spec (the prompt): comp_id PipelineMapvisual pipelineshape octaground whitetreatment wireframemotion kinetic-buildpower throwinstrument laser-trace→Step: prep, listen, clipcurate stages, steps
9s9
matches intent
shared field
signature-3d		resolve56.9–64.6scoalescenceicon·color♪ bed_out
The point of the prep is not to sound smart. It is to be present enough to catch the truth.
on-screen: Use AI to be more present
expected on screen: white ground · octa hero in the shared Signal Field · Lumen leads · coalescence · coalescence · icon·color logo · caption bottom-left
spec (the prompt): comp_id shared Signal Fieldvisual coalescenceshape octaground whitetreatment colormotion coalescencepower coalescenceinstrument coalescence→Use AI to be more present

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)

tiktokUse AI to prep your interview, so you can do the one thing it can't: listen. (AI-assisted) Feed the model the guest's past talks and writing, ask for questions nobody else asks, then run the transcript to mark the three most quotable moments. Be present, catch the surprise.
instagramLet AI prep so you can listen. Feed it their past talks and writing. Ask for questions nobody else asks. Mark the quotable clips from the transcript. #interviews #podcasting #contentcreation #aitools #storytelling
linkedinA model can prep a whole interview for you, which frees you to do the one thing it can't: listen. The move: feed the model the guest's past talks and writing, and ask for the questions nobody else asks. Sharper questions pull realer answers, because the guest can tell you did the homework. Afterward, run the recording through a transcript and have the model mark the three most quotable moments, so your edit drops from a day to an afternoon. The point of the prep isn't to sound smart. It's to be present enough to catch the truth.
xUse AI to prep the interview, so you can do the one thing it can't: listen. Feed it their past work, ask for questions nobody else asks, then mark the quotable clips from the transcript. Be present, catch the surprise. temerarii.com
facebookLet AI prep your interview so you can do the one thing it can't: listen. Feed the model the guest's past talks and writing, ask for questions nobody else asks, then run the transcript to mark the most quotable moments. The point is to be present enough to catch the truth. temerarii.com
threadsUse AI to prep the interview, so you can do the one thing it can't: listen. Feed it their past talks and writing, ask for the questions nobody else asks, then mark the quotable clips from the transcript. Be present, catch the surprise.
pinterestInterview prep with AI: feed the model your guest's past talks and writing, ask for questions nobody else asks, then mark the most quotable moments from the transcript. Podcast interview prep, better questions, transcript clipping, content creation workflow.
blueskyUse AI to prep the interview, so you can listen. Feed it their past work, ask for questions nobody else asks, then mark the quotable clips from the transcript. Be present, catch the surprise. temerarii.com
youtubeBetter Interviews: Let AI Prep So You Can Listen A model can prep a whole interview for you, which frees you to do the one thing it can't: listen. This short shows the move: feed the model the guest's past talks and writing, and ask for the questions nobody else asks. Sharper questions pull realer answers. Afterward, run the recording through a transcript and have the model mark the most quotable moments, so your edit drops from a day to an afternoon. The point of prep isn't to sound smart. It's to be present enough to catch the truth. #interviews #podcasting #contentcreation

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

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