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-W24-Mon-3kind threadweek W24date 2026-06-15campaign thread · Monpillar emerging_techbeat Monasset videoduration 58.7sground redscenes 9

Checklist the per-video bar — engine/sim

96.5/100
plain languagevo coverageno dead airuniquenesscaption fitcompletenesscleanliness
quantitative quality · weights learn from your reviews (engine.sim.memory review social-W24-Mon-3 good|bad)
⚠ 2 flag(s) — not yet ship-ready: copy_generictoo_complex · 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 · 58.7s · comp_id + rendered still + tier + the script

#Layer (comp_id · still · tier)BeatTimecodeMotionLogoAudioVO / on-screen / caption
1s1
matches intent
shared field
signature-3d		open0–6.6sspatial-parallaxicon·liquid-chrome♪ bed_in
AI is a tool, not a wizard, and treating it like one is how teams get burned.
on-screen: AI is a tool, not a wizard
expected on screen: red ground · cone hero in the shared Signal Field · Augur leads · node-graph · spatial-parallax · icon·liquid-chrome logo · caption bottom-left
spec (the prompt): comp_id shared Signal Fieldvisual node-graphshape coneground redtreatment liquid-chromemotion spatial-parallaxpower summoninstrument summon→AI is a tool, not a wizard
2s2
matches intent
shared field
signature-3d		hook6.6–12.5skinetic-buildicon·white-knockout♪ node_lock
Here is what The Big T-M sees as genuinely possible once you drop the hype.
on-screen: What is now actually possible
expected on screen: red ground · cone hero in the shared Signal Field · Augur leads · mark · kinetic-build · icon·white-knockout logo · caption bottom-left
spec (the prompt): comp_id shared Signal Fieldvisual markshape coneground redtreatment white-knockoutmotion kinetic-buildpower laser-lockinstrument laser-trace→What is now actually possible
3s3
matches intent
NumberedList
template		teach12.5–20.2skinetic-buildicon·liquid-chrome♪ node_lock
The shift is grounding: hand the model your own documents so it answers from your truth, not the internet's average.
on-screen: Give the model your real data
expected on screen: red ground · a NumberedList panel over a dimmed Signal Field · Augur leads · node-graph · kinetic-build · icon·liquid-chrome logo · caption bottom-left
spec (the prompt): comp_id NumberedListvisual node-graphshape coneground redtreatment liquid-chromemotion kinetic-buildpower morphinstrument morph+laser→Give the model your real datacurate items, nodes
4s4
matches intent
RankList
template		proof20.2–27.2sreceipts-counticon·white-knockout♪ node_lock
The proof is the citation: a grounded answer points to the source so a human can check it.
on-screen: It cites, you verify
expected on screen: red ground · a RankList panel over a dimmed Signal Field · Augur leads · receipts · receipts-count · icon·white-knockout logo · caption bottom-left
spec (the prompt): comp_id RankListvisual receiptsshape coneground redtreatment white-knockoutmotion receipts-countpower receiptsinstrument spotlight→It cites, you verifycurate rows, statsLabels
5s5
matches intent
LogStream
template		diff27.2–33.5scrossfade-8ficon·liquid-chrome♪ node_lock
The old way was one giant chatbot trying to know everything and confidently making things up.
on-screen: Old way: one giant chatbot
expected on screen: red ground · a LogStream panel over a dimmed Signal Field · Augur leads · code · crossfade-8f · icon·liquid-chrome logo · caption bottom-left
spec (the prompt): comp_id LogStreamvisual codeshape coneground redtreatment liquid-chromemotion crossfade-8fpower morphinstrument morph→Old way: one giant chatbotcurate codeLines, rows
6s6
matches intent
ChecklistCard
template		teach33.5–41.2skinetic-buildicon·white-knockout♪ node_lock
Now you wire small agents, each with one job and one tool, and let them hand off like a relay.
on-screen: Many small agents, one job each
expected on screen: red ground · a ChecklistCard panel over a dimmed Signal Field · Augur leads · node-graph · kinetic-build · icon·white-knockout logo · caption bottom-left
spec (the prompt): comp_id ChecklistCardvisual node-graphshape coneground redtreatment white-knockoutmotion kinetic-buildpower morphinstrument morph+laser→Many small agents, one job eachcurate items, nodes
7s7
matches intent
StatScoreboard
template		proof41.2–47.800000000000004sreceipts-counticon·liquid-chrome♪ node_lock
The receipt is reliability: a model with a narrow job and real data rarely invents an answer.
on-screen: Narrow jobs rarely hallucinate
expected on screen: red ground · a StatScoreboard panel over a dimmed Signal Field · Augur leads · receipts · receipts-count · icon·liquid-chrome logo · caption bottom-left
spec (the prompt): comp_id StatScoreboardvisual receiptsshape coneground redtreatment liquid-chromemotion receipts-countpower receiptsinstrument spotlight→Narrow jobs rarely hallucinatecurate pillar, stats, statsLabels
8s8
matches intent
BlueprintGrid
template		step47.8–53.699999999999996skinetic-buildicon·white-knockout♪ node_lock
Start with one painful workflow, automate it end to end, then copy the pattern outward.
on-screen: Start with one painful workflow
expected on screen: red ground · a BlueprintGrid panel over a dimmed Signal Field · Augur leads · pipeline · kinetic-build · icon·white-knockout logo · caption bottom-left
spec (the prompt): comp_id BlueprintGridvisual pipelineshape coneground redtreatment white-knockoutmotion kinetic-buildpower throwinstrument laser-trace→Start with one painful workflowcurate rows, stages, steps
9s9
matches intent
shared field
signature-3d		resolve53.7–58.7scoalescenceicon·liquid-chrome♪ bed_out
The future that lasts is boring and useful, not flashy and fragile.
on-screen: Boring and useful beats flashy
expected on screen: red ground · cone hero in the shared Signal Field · Augur leads · coalescence · coalescence · icon·liquid-chrome logo · caption bottom-left
spec (the prompt): comp_id shared Signal Fieldvisual coalescenceshape coneground redtreatment liquid-chromemotion coalescencepower coalescenceinstrument coalescence→Boring and useful beats flashy

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)

tiktokAI is a tool, not a wizard. AI-assisted: ground the model in YOUR documents so it answers from your truth and cites sources, then wire small single-job agents that hand off like a relay. Narrow jobs rarely hallucinate. Method below.
instagramAI is a tool, not a wizard. Ground it in your data. Make it cite sources. Wire small single-job agents. Start with one painful workflow. #artificialintelligence #aiagents #automation #futuretech #temerarii
linkedinAI is a tool, not a wizard, and treating it like one is how teams get burned. What's genuinely possible once you drop the hype: ground the model in your own documents so it answers from your truth and cites the source, then wire small agents that each do one job with one tool and hand off like a relay. A narrow job plus real data rarely hallucinates. Takeaway: start with one painful workflow, automate it end to end, then copy the pattern.
xAI is a tool, not a wizard. Ground it in your docs → make it cite → wire small single-job agents that hand off like a relay. Narrow jobs rarely hallucinate. temerarii.com
facebookAI is a tool, not a wizard, and treating it like one is how teams get burned. Here's what's actually possible: ground the model in your own documents so it answers from your truth and cites the source, then wire small agents that each do one job and hand off like a relay. Start with one painful workflow and copy the pattern outward. Want the build? It's on the site.
threadsAI is a tool, not a wizard. Ground it in your own docs so it cites sources, wire small single-job agents that hand off like a relay, start with one painful workflow. Narrow jobs rarely hallucinate.
pinterestArtificial intelligence for business: how to ground models in your own data, get cited answers, build small single-job AI agents, and automate one workflow at a time. Plain-language guide to practical AI without the hype.
blueskyAI is a tool, not a wizard. Ground it in your docs, make it cite, wire small single-job agents that hand off like a relay. Narrow jobs rarely hallucinate. temerarii.com
youtubeAI Is A Tool, Not A Wizard: What's Actually Possible Now Treating AI like a wizard is how teams get burned. This walkthrough drops the hype and shows what works: ground the model in your own documents so it answers from your truth and cites the source, then wire small agents that each do one job with one tool and hand off like a relay. Start with one painful workflow and copy the pattern. AI-assisted production. Give-it-away method from The Big T-M. Chapters: tool not wizard · grounding · citations · small agents · start with one workflow.

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

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