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 longform-W45-Frikind longformweek W45date 2026-11-13campaign longform-youtubepillar multimediabeat asset videoduration 125.5sground whitescenes 7

Checklist the per-video bar — engine/sim

100.0/100
plain languagevo coverageno dead airuniquenesscaption fitcompletenesscleanliness
quantitative quality · weights learn from your reviews (engine.sim.memory review longform-W45-Fri good|bad)
✓ all static checks pass — one-focal/scene · tier-by-beat · one-track caption · colorway · cast+shape correct · no banned/fabricated. (audio + visual tiers verify on the rendered finals — Phase 2)

Composition comp · template family · expected output

composition LongFormChaptersfamily / template LongFormChapters
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 7 scenes · 125.5s · comp_id + rendered still + tier + the script

#Layer (comp_id · still · tier)BeatTimecodeMotionLogoAudioVO / on-screen / caption
1s1
matches intent
shared field
signature-3d
open0–18.0sspatial-parallaxicon·color♪ bed_in
Making fresh content every day is a trap. The social engine does not need more ideas, it needs more shapes of one idea. Today is repurposing. By the end you will know how to take a single piece and let a machine spin it into a week of posts.
on-screen: One idea feeds the whole week
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→One idea feeds the whole week
2s2
matches intent
NumberedList
template
teach18.0–33.9skinetic-buildicon·wireframe♪ node_lock
First move. Make one thing of real depth, like this long video. That is your source. Everything this week comes out of it. You are not writing seven times. You are writing once and cutting six more times. The machine does the cutting.
on-screen: Start with one long piece
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→Start with one long piececurate items, nodes
3s3
matches intent
ChecklistCard
template
teach33.9–50.9skinetic-buildicon·wireframe♪ node_lock
Second move. Wire a transcript tool through MCP so the agent reads the spoken words straight from your video. We use a transcription server that returns clean text with timestamps. No retyping. The model now has the full piece as text, ready to reshape into anything.
on-screen: Pull the words out automatically
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 words out automaticallycurate items, nodes
4s4
matches intent
BuildLog
template
teach50.9–70.7skinetic-buildicon·wireframe♪ node_lock
Third move. Ask the model to find the three sharpest lines in the transcript and mark their timestamps. Feed those marks to a video tool, we use a clipping setup, and it cuts three short videos from the one long one. Each strong line becomes a post. The best parts get their own stage.
on-screen: Cut clips at the good lines
expected on screen: white ground · a BuildLog panel over a dimmed Signal Field · Lumen leads · node-graph · kinetic-build · icon·wireframe logo · caption bottom-left
spec (the prompt): comp_id BuildLogvisual node-graphshape octaground whitetreatment wireframemotion kinetic-buildpower morphinstrument morph+laser→Cut clips at the good linescurate lines, nodes
5s5
matches intent
ProcessFlow
template
teach70.7–89.1skinetic-buildicon·wireframe♪ node_lock
Fourth move. The same point reads differently on each app. Have the model rewrite the idea as a tight caption, a longer story, and a thread, matching how people read on each one. One idea, five doors. You approve the set in a single sitting instead of grinding all week.
on-screen: Reshape for each platform
expected on screen: white ground · a ProcessFlow panel over a dimmed Signal Field · Lumen leads · node-graph · kinetic-build · icon·wireframe logo · caption bottom-left
spec (the prompt): comp_id ProcessFlowvisual node-graphshape octaground whitetreatment wireframemotion kinetic-buildpower morphinstrument morph+laser→Reshape for each platformcurate nodes, steps
6s6
matches intent
StatScoreboard
template
proof89.1–107.5sreceipts-counticon·wireframe♪ node_lock
Honest proof. This exact video is the source for our own Friday. The clips, the captions, the thread you will see from us this week all came out of these words. One piece, many shapes. You are watching the source of a week of posts as it is being recorded.
on-screen: This video is our source
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→This video is our sourcecurate pillar, stats, statsLabels
7s7
matches intent
shared field
signature-3d
resolve107.5–125.5scoalescenceicon·color♪ bed_out
That is repurposing. One deep piece, an automatic transcript, clips at the best lines, a reshape for each platform. You make once and post for days. See the full engine work in public at office dot temerarii dot xyz, then turn your next long piece into a whole week.
on-screen: Repurpose yours at the office
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→Repurpose yours at the office

Format stack 1 aspects · same scenes[], re-cropped

16:9
1920×1080
X/Twitter · YouTube · LinkedIn video

Channels 2 destinations

YouTubeBlog

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

youtubeTurn One Long Video Into a Week of Posts With Automatic Transcripts and Clips Making fresh content every day is a trap. The engine does not need more ideas, it needs more shapes of one idea. In this video we show how to take a single piece and let a machine spin it into a week of posts. The moves: - Start with one long piece. Make one thing of real depth, like a long video. That is your source. Everything this week comes out of it. You write once and cut six more times, and the machine does the cutting. - Pull the words out automatically. Wire a transcript tool through MCP so the agent reads the spoken words straight from your video. We use a transcription server that returns clean text with timestamps. No retyping. - Cut clips at the good lines. Ask the model to find the three sharpest lines in the transcript and mark their timestamps. Feed those marks to a clipping tool and it cuts three short videos from the one long one. - Reshape for each platform. Have the model rewrite the idea as a tight caption, a longer story, and a thread, matching how people read on each app. One idea, five doors. The honest proof: this exact video is the source for our own Friday. The clips, the captions, and the thread you will see from us this week all came out of these words. You are watching the source of a week of posts as it is being recorded. The takeaway: make once and post for days. One deep piece, an automatic transcript, clips at the best lines, a reshape for each platform. See the full engine work in public at office.temerarii.xyz. Keywords: content repurposing, transcript automation, video clipping, MCP, social media.

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