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-W25-Tuekind longformweek W25date 2026-06-23campaign longform-youtubepillar socialbeat asset videoduration 197.2sground blackscenes 8

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-W25-Tue 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 8 scenes · 197.2s · comp_id + rendered still + tier + the script

#Layer (comp_id · still · tier)BeatTimecodeMotionLogoAudioVO / on-screen / caption
1s1
matches intent
shared field
signature-3d
open0–25.5sspatial-parallaxicon·ember-fill♪ bed_in
Today is creative conception, the part everyone treats as a mystery. We are going to kill that. We point the tools at ourselves first and show how we turn a blank week into real ideas, on purpose, with steps you can repeat. By the end you will see the exact way we go from nothing to a concept we can build. No waiting for inspiration, just a method that runs.
on-screen: Where the idea comes from
expected on screen: black ground · torus hero in the shared Signal Field · Nuntius leads · node-graph · spatial-parallax · icon·ember-fill logo · caption bottom-left
spec (the prompt): comp_id shared Signal Fieldvisual node-graphshape torusground blacktreatment ember-fillmotion spatial-parallaxpower summoninstrument summon→Where the idea comes from
2s2
matches intent
NumberedList
template
teach25.5–49.6skinetic-buildicon·wireframe♪ node_lock
Every concept starts from a short written brief: the week's headline, the one service we are teaching, and who it is for. We keep these briefs as a single data file the machine reads. Today's brief is this exact video. We do not brainstorm in our heads, we write the constraints down first, because a tight constraint is what makes an idea sharp instead of mush.
on-screen: Start from a written brief, not a
expected on screen: black ground · a NumberedList panel over a dimmed Signal Field · Nuntius leads · node-graph · kinetic-build · icon·wireframe logo · caption bottom-left
spec (the prompt): comp_id NumberedListvisual node-graphshape torusground blacktreatment wireframemotion kinetic-buildpower morphinstrument morph+laser→Start from a written brief, not a vibecurate items, nodes
3s3
matches intent
ChecklistCard
template
teach49.6–75.1skinetic-buildicon·wireframe♪ node_lock
With the brief in context, we ask Claude Code for a stack of angles, twenty rough directions in a minute. The model is good at width, bad at taste. So we use it for width only. We read all twenty, kill the obvious ones, and keep the two that feel specific to us. The move is not let the AI decide. The move is let it flood, then you judge.
on-screen: Use the model to widen, then you
expected on screen: black ground · a ChecklistCard panel over a dimmed Signal Field · Nuntius leads · node-graph · kinetic-build · icon·wireframe logo · caption bottom-left
spec (the prompt): comp_id ChecklistCardvisual node-graphshape torusground blacktreatment wireframemotion kinetic-buildpower morphinstrument morph+laser→Use the model to widen, then you cutcurate items, nodes
4s4
matches intent
StepFlow
template
teach75.1–99.6skinetic-buildicon·wireframe♪ node_lock
We refuse ideas that float. Each concept has to teach one true service we actually sell, named plainly. If an angle does not map to a real thing on our list, we cut it, no matter how clever. We check this against our services file. That one rule keeps our creative from becoming pretty noise. An idea that does not point at the work is just decoration.
on-screen: Tie every idea to one real service
expected on screen: black ground · a StepFlow panel over a dimmed Signal Field · Nuntius leads · node-graph · kinetic-build · icon·wireframe logo · caption bottom-left
spec (the prompt): comp_id StepFlowvisual node-graphshape torusground blacktreatment wireframemotion kinetic-buildpower morphinstrument morph+laser→Tie every idea to one real servicecurate nodes, steps
5s5
matches intent
BlueprintGrid
template
teach99.6–125.89999999999999skinetic-buildicon·wireframe♪ node_lock
Then we attach a visual idea so the concept can be built, not just described. We have a small set of named shapes and characters in code, and each concept gets matched to one that fits its meaning. The model proposes the match, we approve it. Now the idea has a body. A concept you cannot picture is a concept you cannot ship, so we force a shape before we move on.
on-screen: Give the concept a shape and a
expected on screen: black ground · a BlueprintGrid panel over a dimmed Signal Field · Nuntius leads · node-graph · kinetic-build · icon·wireframe logo · caption bottom-left
spec (the prompt): comp_id BlueprintGridvisual node-graphshape torusground blacktreatment wireframemotion kinetic-buildpower morphinstrument morph+laser→Give the concept a shape and a charactercurate nodes, rows
6s6
matches intent
DiffCard
template
teach125.9–149.3skinetic-buildicon·wireframe♪ node_lock
Before a concept survives, we run it through the same brand check as everything else: plain language, no banned words, specific not generic. Many clever ideas die here because they only sound smart. We would rather lose a concept now than ship a vague one. The test is mechanical on purpose, so our own ego does not get a vote it has not earned.
on-screen: Pressure test it against the brand file
expected on screen: black ground · a DiffCard panel over a dimmed Signal Field · Nuntius leads · node-graph · kinetic-build · icon·wireframe logo · caption bottom-left
spec (the prompt): comp_id DiffCardvisual node-graphshape torusground blacktreatment wireframemotion kinetic-buildpower morphinstrument morph+laser→Pressure test it against the brand filecurate lines, nodes
7s7
matches intent
ComparisonTable
template
proof149.3–173.10000000000002sreceipts-counticon·wireframe♪ node_lock
The honest proof is the thing you are watching. This concept, this order, these beats, came out of that exact pipeline, brief to angles to shape to check. We did not hand carve it and pretend. It was conceived by the method and approved by a human. You can read the brief it came from, and the others, live at office dot temerarii dot xyz.
on-screen: This script is the proof
expected on screen: black ground · a ComparisonTable panel over a dimmed Signal Field · Nuntius leads · receipts · receipts-count · icon·wireframe logo · caption bottom-left
spec (the prompt): comp_id ComparisonTablevisual receiptsshape torusground blacktreatment wireframemotion receipts-countpower receiptsinstrument spotlight→This script is the proofcurate colA, colB, rows, statsLabels
8s8
matches intent
shared field
signature-3d
resolve173.1–197.2scoalescenceicon·ember-fill♪ bed_out
So creative conception, demystified. Write a tight brief. Flood with the model, cut with your taste. Force every idea onto a real service. Give it a shape so it can be built. Test it like everything else. That is ideas on a schedule instead of by luck. Go see the briefs running live at office dot temerarii dot xyz, and start writing your own constraints first.
on-screen: Idea on demand, not by luck
expected on screen: black ground · torus hero in the shared Signal Field · Nuntius leads · coalescence · coalescence · icon·ember-fill logo · caption bottom-left
spec (the prompt): comp_id shared Signal Fieldvisual coalescenceshape torusground blacktreatment ember-fillmotion coalescencepower coalescenceinstrument coalescence→Idea on demand, not by luck

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)

youtubeWhere the Idea Comes From: A Written Brief, Not a Lucky Mood Good ideas are not a mood you wait for. This video shows where the idea actually comes from: a written brief, widened by a model, then narrowed by you. What you will learn: - Start from a written brief, not a blank page: name the audience, the constraint, and the job before you ask for anything. - Use the model to widen, then you narrow: let it spread the option set wide, then you cut it down with judgment the model does not have. - Tie every idea to one real service: an idea that does not map to something you actually offer is a distraction, so anchor each one. - Give the concept a shape and a hook, then pressure-test it against the brand file so the weak ones fail before they cost time. Then the proof: this very script is the proof. It came out of that exact method, an idea on demand, anchored to a service, checked against the brand, not luck. Keywords: content ideation, creative brief, brand alignment, idea generation, content strategy. See the whole system run in public at temerarii.xyz.

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