every launch video on the internet is a template now. left is Gojiberry AI launch video. right is the version I had Opus 5.5 rebuild for my brand in 15 minutes. same motion, same cuts, playing in sync. I’m open sourcing the exact prompt + process so you can... recreate any launch video on the internet. full prompt below ↓show more

TASK: frame-locked 1:1 remake of REF=[path/to/reference.mp4]. Swap ONLY: brand name→[NAME], logo→[path/to/logo.png|svg], palette→[hex list or "derive from logo"], platform UI→[e.g. LinkedIn→X], faces/screens→[source]. All else identical: layout, sizes, positions, timing, easing, camera, cuts, blur, cursor path, typing cadence, audio beat grid. ACCEPTANCE: REF and remake stacked + frame-locked → same pose every frame. Cuts 0 frames off. Position/size error ≤1% of frame. Typing char-count, cursor, camera on the same frames. Differences allowed only where the swap forces them (word widths); report each. PHASE 0 — ANALYSIS (no building yet) - ffprobe fps/res/duration. Extract ALL frames 0-based to ref/full/fNNNN.jpg + audio to ref/audio.wav. 1fps contact sheets for overview. - Detect cuts via per-frame mean-abs-diff spikes; confirm visually. Write SPEC.md: shot table (id, f0–f1, content, transition in/out), per-shot detail, component inventory, verbatim on-screen text in order, colour tokens sampled from pixels, font sizes from cap height, cursor paths (tip xy per frame), camera keyframes. - MEASURE with numpy on ref frames (ink bounding boxes, flood fills, template match), never eyeball. Store per-frame sample arrays for every major move; the spec's prose is a guide, ref/full is truth. - Audio: STT with word timestamps (narration table: line, start, end, text). Music BPM + beat phase (onset autocorrelation), drop time, loudness arc per section, hard-stop time. SFX hit times from onset/spectral analysis. Voice pitch/wpm. PHASE 1 — ENGINE (you, before any agents) - Single HTML page, stage at REF native resolution. renders frame F=t*fps as a PURE function of F: no timers, no Date, no Math.random (seeded hash), no CSS transitions/animations. Shots register SHOT({id,f0,f1,render(lf,F)}) returning HTML; seek routes F to its shot. window.ready=true only after document.fonts loaded + all images decoded. - core.js shared helpers: easing set, kf(F,keys,ease), samples(F,f0,arr) interpolating measured arrays, camera(inner,scale,tx,ty,origin,blur), directional motion blur (SVG feGaussianBlur), cursor (glyph MEASURED from ref, press-scale curve measured), ripple, text reveal, brand tokens, logo component (mask-image of logo so any fill/gradient works), avatar/persona pickers. - Palette swap as ONE deterministic filter applied to each rendered HTML string (hex/rgb/rgba re-hued by band, lightness preserved) so hard-coded colours in shots can't leak the old brand. Bitmaps untouched. - render.mjs (Playwright Chromium, deviceScaleFactor 1): modes stills <frames> | compare <frames> (ref left | ours right + labelled sheet) | full <f0> <f1>. Per-agent OUT dirs so parallel runs don't collide. Print page errors. - REF over remake stacked, frame-locked mp4. PNG frames → h264 at REF fps, mux audio. PHASE 2 — PARALLEL BUILD - Split shots into 4 contiguous groups; one agent each; each writes ONLY shots/<G>.js (IIFE, helpers prefixed <G>_), never edits core.js (request changes from you). Give each: BRIEF.md (acceptance bar, swap rules, file rules, verify loop), SPEC.md sections, core.js API. - Verify loop per shot: compare first/last frame, every keyframe, 2 frames into each transition; iterate until within tolerance. ≤15 frames per render call; one render at a time per agent. - Report table: shot | frames | MATCHES/CLOSE/ROUGH | residual diff | frames that would drift when stacked | spec errors found vs ref. - 5th agent = audio: royalty-free commercial-OK track (record URL+licence), time-stretch to REF BPM, cut on beats so drop/breaks/silences land on REF times; SFX synthesized (numpy) on REF hit times; mix script with VO slot table (REF start times, text swapped) that fits each line (≤8% stretch), ducks music under voice, loudness-normalizes (~-14 LUFS, TP ≤ -1). Never reuse REF music or voice. PHASE 3 — INTEGRATE - Unify shared glyphs/components across groups (cursor, DM window, logo) — one definition, everyone uses it. - Full render all frames (2–3 parallel chunks max), encode with mix, run build 1-per-second ref|ours contact sheet + sheets at every group seam (last 2 / first 2 frames). Inspect. Fix drift. Re-render. Scan all frames for leftover old-brand colour. - VO: generate per line (any TTS), trim silence, place on slots, re-mix. PITFALLS (seen in practice) - Measure text widths only AFTER fonts load (measureText before load caches fallback widths → words collide). Lazy-init any width tables. - Word-length changes from the brand swap shift centred layouts; keep shared elements on ref positions, absorb the delta in the swapped word, report it. - Randomised bursts/particles won't match stacked; drive the visible ones from measured tracks. - Ref whips may be crisp (no blur); check before adding motion blur. - Headless Chromium may need to run outside any OS sandbox; use swiftshader/ANGLE if WebGL is involved. - Never claim a shot matches without viewing ref|ours side by side for it. DELIVER: remake.mp4 (with audio), sync-check.mp4, SPEC.md, source, and a list of every remaining difference vs REF with frame numbers.

cooking even harder.

"sahus" lmaoo I run out of elvevenlabs credits so claude used some random opensource voice model 😂

@gojiberryai I didn’t believe this created by AI crazyyyy

@gojiberryai try it out gang

@robj3d3 oops i forgot to get permission to use u for the vid 😬

@gojiberryai Dang, I’ll try that for my next one. Opus made this with remotion, blender, hyperframes, MCP, and my Hermes agent’s context:

@gojiberryai Im not at all against AI. but launch videos, marketing was never about how good pixels look on screen. If you want your brand to be perceived as a cheap product and knock off of the real one thats why you re doing with this videos. The tech is cool but this is lame

@gojiberryai Thanks for sharing this!

@gojiberryai WTF it's so perfect

@gojiberryai 15‑minute rebuild, same cuts, flawless sync. token cost negligible, this is how launch vids should be

@gojiberryai how much the whole video costed?

@gojiberryai like a few 100k tokens. basically free gang

@gojiberryai watched cloned launch videos break on revision two once product footage changed

Do you need to install some specific skill beforehand?

I had my own skill. I’ll be sharing it soon. :)

Yes please, thanks

@gojiberryai thanks for sharing

@gojiberryai I still cannot wow enough how this is game changing. Thanks for sharing, I tried and it worked. I even managed to improve it further and create creative videos, not just clones.

@gojiberryai did u add anything like images and stuff

@gojiberryai nah just opus 5.5. ima launch a skill soon

@gojiberryai can't wait

@gojiberryai if you gonna make more of these you will probably need something like

@gojiberryai whats that?

@gojiberryai Thanks for sharing. Very useful. I tried to build this one but yours is better

want to make this yourself? exact workflow + guide here &gt;

@gojiberryai MG

@gojiberryai Felt that. Template launch energy is everywhere. Omneky’s Tastebench is how we pressure-test whether a cut still feels distinctive before it ships — same stack, less sameness.