Checklist the per-video bar — engine/sim
98.0/100
quantitative quality · weights learn from your reviews (
engine.sim.memory review longform-W38-Fri good|bad)⚠ 1 flag(s) — not yet ship-ready: copy_generic · see docs/strategy/VIDEO-CHECKLIST.md
Composition comp · template family · expected output
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 6 scenes · 116.8s · comp_id + rendered still + tier + the script
| # | Layer (comp_id · still · tier) | Beat | Timecode | Motion | Logo | Audio | VO / on-screen / caption |
|---|---|---|---|---|---|---|---|
| 1 |  matches intent shared field signature-3d | open | 0–18.4s | spatial-parallax | icon·white-knockout | ♪ bed_in | Today is copywriting, the words that do the selling. Stop guessing, start printing. By the end you will know how to use a model as a fast writing partner that drafts from your real proof, so your pages stop sounding like everyone else and start sounding like a true offer. on-screen: Copy that sells, written with you |
expected on screen: red ground · icosa hero in the shared Signal Field · Signum leads · node-graph · spatial-parallax · icon·white-knockout logo · caption bottom-left spec (the prompt): comp_id shared Signal Fieldvisual node-graphshape icosaground redtreatment white-knockoutmotion spatial-parallaxpower summoninstrument summon→Copy that sells, written with you | |||||||
| 2 |  matches intent NumberedList template | teach | 18.4–39.7s | kinetic-build | icon·white-knockout | ♪ node_lock | First, gather your raw material: real lines from happy customers, the actual problem you solve, the plain facts of the offer. Hand all of it to the model. The mistake is asking it to write from nothing, so it fills the gap with the empty hype words everyone reaches for. Give it truth and it writes truth back. on-screen: Feed it proof, not adjectives |
expected on screen: red ground · a NumberedList panel over a dimmed Signal Field · Signum leads · node-graph · kinetic-build · icon·white-knockout logo · caption bottom-left spec (the prompt): comp_id NumberedListvisual node-graphshape icosaground redtreatment white-knockoutmotion kinetic-buildpower morphinstrument morph+laser→Feed it proof, not adjectivescurate items, nodes | |||||||
| 3 |  matches intent ChecklistCard template | teach | 39.7–60.2s | kinetic-build | icon·liquid-chrome | ♪ node_lock | Ask the agent to list the top three reasons a buyer would say no, then write a line that answers each one. This is the work most pages skip. People do not need more excitement, they need their worry handled. The model is good at naming the quiet objection. You make sure the answer is honest. on-screen: Write the objection, then answer it |
expected on screen: red ground · a ChecklistCard panel over a dimmed Signal Field · Signum leads · node-graph · kinetic-build · icon·liquid-chrome logo · caption bottom-left spec (the prompt): comp_id ChecklistCardvisual node-graphshape icosaground redtreatment liquid-chromemotion kinetic-buildpower morphinstrument morph+laser→Write the objection, then answer itcurate items, nodes | |||||||
| 4 |  matches intent WireframeMock template | teach | 60.2–80.7s | kinetic-build | icon·white-knockout | ♪ node_lock | Now run an edit pass. Tell the model to flag any sentence that promises more than you can prove or leans on a tired hype phrase, and rewrite it plain. Read each one aloud. If it sounds like a brochure, it goes. The goal is copy an eleven-year-old understands and a skeptic believes. Short, concrete, true. on-screen: Cut every line that lies a little |
expected on screen: red ground · a WireframeMock panel over a dimmed Signal Field · Signum leads · node-graph · kinetic-build · icon·white-knockout logo · caption bottom-left spec (the prompt): comp_id WireframeMockvisual node-graphshape icosaground redtreatment white-knockoutmotion kinetic-buildpower morphinstrument morph+laser→Cut every line that lies a littlecurate nodes | |||||||
| 5 |  matches intent StatScoreboard template | proof | 80.7–100.9s | receipts-count | icon·liquid-chrome | ♪ node_lock | Every word on our own site is built this way, from the real method and real results, drafted with a model and cut until it is plain. We do not write that the studio is amazing. We show you the machine running. You can read it, and see how the copy was made, at office.temerarii.xyz. on-screen: We write ours the same way |
expected on screen: red ground · a StatScoreboard panel over a dimmed Signal Field · Signum leads · receipts · receipts-count · icon·liquid-chrome logo · caption bottom-left spec (the prompt): comp_id StatScoreboardvisual receiptsshape icosaground redtreatment liquid-chromemotion receipts-countpower receiptsinstrument spotlight→We write ours the same waycurate pillar, stats, statsLabels | |||||||
| 6 |  matches intent shared field signature-3d | resolve | 100.9–116.80000000000001s | coalescence | icon·white-knockout | ♪ bed_out | So: feed the model your real proof, answer the buyer's no, and cut every line that overpromises. Copy that is plain and true outsells copy that shouts. The whole approach is live at office.temerarii.xyz. Rewrite one page this week, starting with the objections. on-screen: Say it plain, prove it true |
expected on screen: red ground · icosa hero in the shared Signal Field · Signum leads · coalescence · coalescence · icon·white-knockout logo · caption bottom-left spec (the prompt): comp_id shared Signal Fieldvisual coalescenceshape icosaground redtreatment white-knockoutmotion coalescencepower coalescenceinstrument coalescence→Say it plain, prove it true | |||||||
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)
youtubeWrite copy that sells by feeding the model real proof, not adjectives
Most pages sound like everyone else because they are written from nothing. This is how to use a model as a fast writing partner that drafts from your real proof, so your copy reads like a true offer.
The method:
- Feed it proof, not adjectives. Gather real lines from happy customers, the actual problem you solve, and the plain facts of the offer. Give the model truth and it writes truth back. Ask it to write from a blank prompt and it fills the gap with empty hype.
- Write the objection, then answer it. Ask the agent for the top three reasons a buyer would say no, then write an honest line for each. People do not need more excitement; they need their worry handled.
- Cut every line that lies a little. Run an edit pass: flag any sentence that promises more than you can prove or leans on a tired phrase, and rewrite it plain. Read each one aloud. If it sounds like a brochure, it goes.
The honest proof: every word on our own site is built this way, from the real method and real results, drafted with a model and cut until it is plain. We do not write that the studio is amazing; we show the machine running.
Feed the model your real proof, answer the buyer's no, and cut every line that overpromises. Rewrite one page this week, starting with the objections. The whole approach is at office.temerarii.xyz.
Keywords: copywriting with AI, objection handling, plain language copy, sales pages.