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-Sun-4kind threadweek W33date 2026-08-16campaign thread · Sunpillar strategic_relationsbeat Sunasset videoduration 61.3sground blackscenes 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-Sun-4 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 · 61.3s · comp_id + rendered still + tier + the script

#Layer (comp_id · still · tier)BeatTimecodeMotionLogoAudioVO / on-screen / caption
1s1
matches intent
shared field
signature-3d		open0–7.0sspatial-parallaxicon·ember-fill♪ bed_in
We are close to a world where a clear product spec writes most of the first version itself.
on-screen: Soon the spec writes the app
expected on screen: black ground · tetra hero in the shared Signal Field · Nexus leads · node-graph · spatial-parallax · icon·ember-fill logo · caption bottom-left
spec (the prompt): comp_id shared Signal Fieldvisual node-graphshape tetraground blacktreatment ember-fillmotion spatial-parallaxpower summoninstrument summon→Soon the spec writes the app
2s2
matches intent
shared field
signature-3d		hook7.0–13.3skinetic-buildicon·wireframe♪ node_lock
Old way, the product manager wrote a long doc that engineering skimmed and then quietly ignored.
on-screen: The old way: a doc nobody read
expected on screen: black ground · tetra hero in the shared Signal Field · Nexus leads · mark · kinetic-build · icon·wireframe logo · caption bottom-left
spec (the prompt): comp_id shared Signal Fieldvisual markshape tetraground blacktreatment wireframemotion kinetic-buildpower laser-lockinstrument laser-trace→The old way: a doc nobody read
3s3
matches intent
NumberedList
template		teach13.3–20.6skinetic-buildicon·wireframe♪ node_lock
Here is the move: write your spec as structured data, with goals, rules, and edge cases as plain fields.
on-screen: Write the spec as data
expected on screen: black ground · a NumberedList panel over a dimmed Signal Field · Nexus leads · node-graph · kinetic-build · icon·wireframe logo · caption bottom-left
spec (the prompt): comp_id NumberedListvisual node-graphshape tetraground blacktreatment wireframemotion kinetic-buildpower morphinstrument morph+laser→Write the spec as datacurate items, nodes
4s4
matches intent
ComparisonTable
template		proof20.6–27.200000000000003sreceipts-counticon·wireframe♪ node_lock
Now a model can read that file and scaffold the screens, tests, and tickets straight from it.
on-screen: Machine reads it, builds the stub
expected on screen: black ground · a ComparisonTable panel over a dimmed Signal Field · Nexus leads · receipts · receipts-count · icon·wireframe logo · caption bottom-left
spec (the prompt): comp_id ComparisonTablevisual receiptsshape tetraground blacktreatment wireframemotion receipts-countpower receiptsinstrument spotlight→Machine reads it, builds the stubcurate colA, colB, rows, statsLabels
5s5
matches intent
JsonDiff
template		diff27.2–34.5scrossfade-8ficon·wireframe♪ node_lock
A doc that only humans can read is a wish. A spec a machine can read is a plan.
on-screen: A spec a machine can read
expected on screen: black ground · a JsonDiff panel over a dimmed Signal Field · Nexus leads · code · crossfade-8f · icon·wireframe logo · caption bottom-left
spec (the prompt): comp_id JsonDiffvisual codeshape tetraground blacktreatment wireframemotion crossfade-8fpower morphinstrument morph→A spec a machine can readcurate codeLines, fileName, lines
6s6
matches intent
ChecklistCard
template		teach34.5–41.1skinetic-buildicon·wireframe♪ node_lock
Add acceptance tests right inside the spec, so done is defined before a single line gets written.
on-screen: Put acceptance tests in the spec
expected on screen: black ground · a ChecklistCard panel over a dimmed Signal Field · Nexus leads · node-graph · kinetic-build · icon·wireframe logo · caption bottom-left
spec (the prompt): comp_id ChecklistCardvisual node-graphshape tetraground blacktreatment wireframemotion kinetic-buildpower morphinstrument morph+laser→Put acceptance tests in the speccurate items, nodes
7s7
matches intent
KpiGrid
template		proof41.1–47.7sreceipts-counticon·wireframe♪ node_lock
The build runs the spec tests, and when they pass, everybody agrees the feature is actually finished.
on-screen: Tests pass, the feature is done
expected on screen: black ground · a KpiGrid panel over a dimmed Signal Field · Nexus leads · receipts · receipts-count · icon·wireframe logo · caption bottom-left
spec (the prompt): comp_id KpiGridvisual receiptsshape tetraground blacktreatment wireframemotion receipts-countpower receiptsinstrument spotlight→Tests pass, the feature is donecurate kpis, statsLabels
8s8
matches intent
ProcessFlow
template		step47.7–54.7skinetic-buildicon·wireframe♪ node_lock
Three steps: write goals and rules as fields, attach acceptance tests, then let the model generate the stub.
on-screen: Step: goals, rules, tests, generate
expected on screen: black ground · a ProcessFlow panel over a dimmed Signal Field · Nexus leads · pipeline · kinetic-build · icon·wireframe logo · caption bottom-left
spec (the prompt): comp_id ProcessFlowvisual pipelineshape tetraground blacktreatment wireframemotion kinetic-buildpower throwinstrument laser-trace→Step: goals, rules, tests, generatecurate stages, steps
9s9
matches intent
shared field
signature-3d		resolve54.7–61.300000000000004scoalescenceicon·ember-fill♪ bed_out
When the spec is the source, the product manager stops writing prose and starts designing the system.
on-screen: The PM becomes the architect
expected on screen: black ground · tetra hero in the shared Signal Field · Nexus leads · coalescence · coalescence · icon·ember-fill logo · caption bottom-left
spec (the prompt): comp_id shared Signal Fieldvisual coalescenceshape tetraground blacktreatment ember-fillmotion coalescencepower coalescenceinstrument coalescence→The PM becomes the architect

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)

tiktokSoon your product spec writes the first version of the app. (AI-assisted) The move: write the spec as STRUCTURED DATA, goals and rules as fields, acceptance tests inside it. A model scaffolds screens, tests, and tickets from the file. A doc humans read is a wish. A spec a machine reads is a plan.
instagramStop writing product docs nobody reads. Write the spec as data: goals, rules, edge cases as fields. Attach acceptance tests. Let the model scaffold the build. #productmanagement #spec #aitools #buildinpublic #softwaredev
linkedinWe're close to a world where a clear product spec writes most of the first version itself. The shift: stop writing long prose docs engineering skims and ignores. Write the spec as structured data, with goals, rules, and edge cases as plain fields, and attach acceptance tests right inside it. Now a model can read that file and scaffold the screens, tests, and tickets directly. A doc only humans can read is a wish. A spec a machine can read is a plan. When the spec is the source of truth, the PM stops writing prose and starts designing the system.
xSoon the spec writes the first version of the app. Write it as structured data: goals, rules, edge cases as fields, acceptance tests inside. A doc humans read is a wish. A spec a machine reads is a plan. temerarii.com
facebookWe're close to product specs that write the first version themselves. The move: write the spec as structured data, with goals and rules as fields and acceptance tests inside, so a model can scaffold screens, tests, and tickets from it. A spec a machine can read is a plan. temerarii.com
threadsSoon a clear product spec writes the first version of the app. Write it as structured data: goals, rules, edge cases as fields, acceptance tests inside. The model scaffolds the build. A doc humans read is a wish. A machine-readable spec is a plan.
pinterestFuture of product management: write specs as structured data with goals, rules, and acceptance tests as fields so AI can scaffold the build. Spec-driven development, machine-readable product spec, AI product management, acceptance test design.
blueskySoon the spec writes the first version of the app. Write it as structured data: goals, rules, acceptance tests as fields. A doc humans read is a wish, a machine-readable spec is a plan. temerarii.com
youtubeProduct Management: Write Specs as Data, Let AI Build We're close to product specs that write the first version of an app themselves. This short shows the move: stop writing long prose docs engineering ignores. Write the spec as structured data, with goals, rules, and edge cases as plain fields, and attach acceptance tests right inside it. A model can then scaffold the screens, tests, and tickets directly from the file. A doc only humans can read is a wish. A spec a machine can read is a plan. The PM becomes the architect. #productmanagement #spec #aitools

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

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