Checklist the per-video bar — engine/sim
100.0/100
quantitative quality · weights learn from your reviews (
engine.sim.memory review longform-W38-Wed 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
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 · 142.3s · 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 | 0–19.1s | spatial-parallax | icon·ember-fill | ♪ bed_in | Today is SEO, but the honest version, where you write for the questions real people already ask you. Stop guessing, start printing. By the end you will know how to pull your own search data and let a model turn it into pages that answer, so you stop chasing words nobody types. on-screen: SEO from your own queries |
expected on screen: white ground · box hero in the shared Signal Field · Faber leads · node-graph · spatial-parallax · icon·ember-fill logo · caption bottom-left spec (the prompt): comp_id shared Signal Fieldvisual node-graphshape boxground whitetreatment ember-fillmotion spatial-parallaxpower summoninstrument summon→SEO from your own queries | |||||||
| 2 |  matches intent NumberedList template | teach | 19.1–40.0s | kinetic-build | icon·color | ♪ node_lock | Start with the data you already own. Wire an MCP into Google Search Console and pull your own query report: the exact phrases people typed before they found you, and where you rank for each. This is the gold most owners never open. It is your audience telling you, in their words, what they want from you. on-screen: Wire in Search Console |
expected on screen: white ground · a NumberedList panel over a dimmed Signal Field · Faber leads · node-graph · kinetic-build · icon·color logo · caption bottom-left spec (the prompt): comp_id NumberedListvisual node-graphshape boxground whitetreatment colormotion kinetic-buildpower morphinstrument morph+laser→Wire in Search Consolecurate items, nodes | |||||||
| 3 |  matches intent ChecklistCard template | teach | 40.0–60.5s | kinetic-build | icon·color | ♪ node_lock | Ask the model to find queries where you sit on page two, ranked just below the top. Those are near misses, real demand you almost catch. Have the agent list them. Each one is a page that needs a small push, not a new site. You fix what is close before you chase what is far. on-screen: Find the near-miss page |
expected on screen: white ground · a ChecklistCard panel over a dimmed Signal Field · Faber leads · node-graph · kinetic-build · icon·color logo · caption bottom-left spec (the prompt): comp_id ChecklistCardvisual node-graphshape boxground whitetreatment colormotion kinetic-buildpower morphinstrument morph+laser→Find the near-miss pagecurate items, nodes | |||||||
| 4 |  matches intent JsonDiff template | teach | 60.5–82.8s | kinetic-build | icon·color | ♪ node_lock | Now feed the agent one near-miss page plus the real queries it almost ranks for, and ask it to rewrite the page to answer those questions plainly. Headings that match how people ask, a short answer up top. Let the model rewrite from your own data, not from a guess about what a search engine wants. The engine wants the answer. on-screen: Let it rewrite from the data |
expected on screen: white ground · a JsonDiff panel over a dimmed Signal Field · Faber leads · node-graph · kinetic-build · icon·color logo · caption bottom-left spec (the prompt): comp_id JsonDiffvisual node-graphshape boxground whitetreatment colormotion kinetic-buildpower morphinstrument morph+laser→Let it rewrite from the datacurate fileName, lines, nodes | |||||||
| 5 |  matches intent StackTrace template | teach | 82.8–104.4s | kinetic-build | icon·color | ♪ node_lock | Add a small block of schema markup, the structured data that tells search engines and AI answer boxes what the page is. Ask the agent to generate it from the page you just wrote. This is how you show up when someone asks an assistant, not just a search bar. The model writes the markup. You paste it in. on-screen: Mark it up so machines read it |
expected on screen: white ground · a StackTrace panel over a dimmed Signal Field · Faber leads · node-graph · kinetic-build · icon·color logo · caption bottom-left spec (the prompt): comp_id StackTracevisual node-graphshape boxground whitetreatment colormotion kinetic-buildpower morphinstrument morph+laser→Mark it up so machines read itcurate errMsg, errType, fix, frames, nodes | |||||||
| 6 |  matches intent KpiGrid template | proof | 104.4–123.5s | receipts-count | icon·color | ♪ node_lock | We do this on our own site. Every long piece is built from the questions people actually ask about running a studio on AI, pulled the same way, then written to answer them straight. You can read the results, and watch how they were made, at office.temerarii.xyz. The method is the marketing. on-screen: Our pages answer real questions |
expected on screen: white ground · a KpiGrid panel over a dimmed Signal Field · Faber leads · receipts · receipts-count · icon·color logo · caption bottom-left spec (the prompt): comp_id KpiGridvisual receiptsshape boxground whitetreatment colormotion receipts-countpower receiptsinstrument spotlight→Our pages answer real questionscurate kpis, statsLabels | |||||||
| 7 |  matches intent shared field signature-3d | resolve | 123.5–142.3s | coalescence | icon·ember-fill | ♪ bed_out | So: pull your own queries from Search Console, find the near misses, let a model rewrite the page to answer them, and mark it up so machines can read it. You stop guessing keywords and start answering people. It is all live at office.temerarii.xyz. Start with one page on page two. on-screen: Answer what they already ask |
expected on screen: white ground · box hero in the shared Signal Field · Faber leads · coalescence · coalescence · icon·ember-fill logo · caption bottom-left spec (the prompt): comp_id shared Signal Fieldvisual coalescenceshape boxground whitetreatment ember-fillmotion coalescencepower coalescenceinstrument coalescence→Answer what they already ask | |||||||
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)
youtubeSEO from your own search data: write pages that answer real questions
This is the honest version of SEO, where you write for the questions real people already ask you, so you stop chasing words nobody types.
The method:
- Wire in Search Console. Connect an MCP into Google Search Console and pull your own query report: the exact phrases people typed before they found you, and where you rank. This is the gold most owners never open.
- Find the near-miss page. Ask the model for queries where you sit on page two, ranked just below the top. Each is a page that needs a small push, not a new site. Fix what is close before you chase what is far.
- Let it rewrite from the data. Feed the agent one near-miss page plus the real queries it almost ranks for, and ask it to rewrite the page to answer those questions plainly, with headings that match how people ask and a short answer up top.
- Mark it up so machines read it. Add a small block of schema, the structured data that tells search engines and AI answer boxes what the page is. The model writes the markup; you paste it in.
The honest proof: we do this on our own site. Every long piece is built from the questions people actually ask about running a studio on AI, then written to answer them straight. The method is the marketing.
Pull your own queries from Search Console, find the near misses, let a model rewrite the page, and mark it up. Start with one page on page two. It is all at office.temerarii.xyz.
Keywords: SEO, Google Search Console, schema markup, near-miss queries, AEO.