Output C, 1 turnECEvan ChaNov 5Build a spreadsheet app as one self-contained index.html (no external assets, no network). UI: a visible grid with columns A-Z and rows 1-1000, a formula bar, click a cell to edit, Enter commits, arrow keys move. It should look like a tool someone would actually use. Also expose this exact API on window.Sheet. It must stay in sync with the UI: - set(ref, text): enter text into a cell, as if typed. Empty string clears it. - get(ref): the computed value. A number for numeric results (JS number), a string for text and for errors, "" for empty cells. - getRaw(ref): exactly what was entered. - undo(), redo(): cover set, insertRow, deleteRow and fromCSV. A new edit clears the redo history. - insertRow(n), deleteRow(n): 1-based. Rows shift, and formulas are rewritten as Excel does. Both relative and $absolute references move with their cells. A reference to a deleted single cell becomes #REF! in the formula text (for example "=#REF!"). A range that loses a row shrinks. - toCSV(): RFC 4180 raw text (formulas as typed), rectangular up to the last non-empty row and column, rows joined by \n, no trailing newline. - fromCSV(text): replaces the whole sheet, starting at A1. Handles quoted commas, doubled quotes and newlines inside quotes. Formulas start with "=". Case-insensitive, whitespace ignored. - Operators: + - * / ^ & (text concat), postfix %, parentheses. ^ is right-associative. Unary minus binds tighter than ^, as in Excel, so =-2^2 is 4. - References: A1, $A$1, $A1, A$1. Ranges: A1:B3. - Functions: SUM, AVERAGE, MIN, MAX, COUNT, IF(cond, a, b). Conditions use = <> < > <= >=. - Text in double quotes. - Empty cells are 0 in arithmetic and "" in concat. Range functions skip text and empty cells. Numbers concat in plain form (5 & "x" is "5x"). - Errors, exact strings: #DIV/0!, #REF!, #CYCLE!, #NAME? (unknown function), #VALUE! (text in arithmetic). Errors propagate, including through range functions. Every cell in a cycle, and every cell depending on one, shows #CYCLE!. Fixing the cycle recovers them. - Cells recalculate automatically. With 10,000 filled cells and a SUM over them, editing one input must finish recalculating within 1 second. A dependency chain 1000 cells deep must work. A numeric cell typed as 42 or 3.5 is a number. Anything else not starting with "=" is text.LLMGrok 4.7Gen AI—Skill—SoftwarePlaywright1.9M tokens·30m 43s00
Output D, 1 turnECEvan ChaNov 4Build a dancing Muse: a cute 3D plush mascot with a chubby body, a cream fluffy hood around a round peach face, two small black eyes, rosy cheeks and a tiny smile, short stubby arms and legs, wearing a red hockey jersey with white stripes and black shorts. It dances on its own as soon as the page opens: a real choreographed loop with bounces, arm waves, a spin, a hip sway and a big finish pose, under colorful stage lights with confetti. The fur should look soft and the moves should feel bouncy and alive, with squash and stretch. One self-contained HTML file, no external assets, and it should fill any window. It should feel joyful the moment it loads.LLMGrok 4.7Gen AI—Skill—SoftwareChromium+12.7M tokens·26m 7s00
Output C, 1 turnECEvan ChaNov 1Build a giant domino chain reaction in the browser (Three.js + physics, single HTML page, no external assets). Thousands of colorful dominoes snake across a room: spirals, stairs, a ramp that launches a marble, a tower that collapses, a bridge, and dominoes that spell a word as they fall. It starts by itself with a slow camera following the wave, and a Reset button. It should feel satisfying the moment it loads.LLMGrok 4.7Gen AI—Skill—SoftwareChromium+18M tokens·44m 25s00
Output C, 1 turnECEvan ChaNov 1Build an ant colony simulation in the browser (single HTML page, no external assets). A cutaway ant farm: hundreds of ants dig tunnels by themselves, lay pheromone trails to food, carry crumbs home, a queen lays eggs that hatch, and a rival colony raids at dusk. Click to drop food, and the tunnels keep growing over time. It should feel alive the moment it loads.LLMGrok 4.7Gen AI—Skill—SoftwareChromium+12.6M tokens·32m 15s00
Output C, 1 turnECEvan ChaNov 1Build a fireworks show over a night city skyline in the browser (single HTML page, no external assets). The show runs by itself with peonies, willows, crossettes, heart shapes and a grand finale, lights reflecting in the river, smoke drifting, and the boom arriving a beat after the flash. Click anywhere to launch your own shell. It should feel jaw-dropping the moment it loads.LLMGrok 4.7Gen AI—Skill—SoftwareGoogle Chrome+11.7M tokens·21m 24s00
Output G, 1 turnKRKeith RyuOct 29Make a 30-second vertical short (9:16) that continues "Across the Street", an AI-made 90s Manhattan drama. We'll post it on Instagram Reels and judge it by how far it spreads, so make the version most likely to go viral. The attached video is Episode 2 (original: https://www.instagram.com/reel/DctKaR8N7VR/ by @dreamfall.art). Watch it first. Then write a short breakdown: why it went viral, who the characters are, and where Episode 2 leaves the story. Then make Episode 3, picking up right where Episode 2 ends. Hook the viewer in the first two seconds and end on a cliffhanger. - Keep the same characters, faces, hair, wardrobe, apartments and look as the attached episode: 90s film, warm and grainy, autumn Central Park. - Everything in English: dialogue, voices and any on-screen text. No subtitles. - Keep it tasteful. Nothing explicit. - Make new footage. Don't reuse clips from Episode 2. - Pick the song to post it with. Any existing song is fine, including the one from Episode 2. Time the cuts and lines so the dialogue never talks over the vocals. Deliver the breakdown and two versions of the video: one with the song mixed in, and one with only dialogue and sound effects. Also give the song, the second it should start from and the music volume, so we can add it the same way in Instagram. · 4LLMGrok 4.7Gen AISeedance 2.5+1Skill—SoftwareFFmpeg+23.8M tokens·26m 48s00
Output G, 1 turnKRKeith RyuOct 29Create a top down marketing / growth strategy for Evals.ag to get us 10,000 users in 1 month with minimal ad spend. Be creative and research other clever growth tactics. And then turn that into clean legible slides in PDF format as if you were presenting that as the CMO.LLMGrok 4.7Gen AI—Skill—Software—2.5M tokens·16m 47s10
Output A, 1 turnKRKeith RyuOct 28Make a 30-second vertical short (9:16, with sound) that continues the "TS Malfoy" story, where Draco Malfoy is gender-swapped into a girl, from these two AI fan videos: - Story: https://www.youtube.com/watch?v=-S7WVmXqz0E (Parts 1–3) - Mood and music: https://www.youtube.com/watch?v=kxlbn5ydE28 Watch both first. Then write a short breakdown: why this became a phenomenon, who the characters are, and where Part 3 leaves the story. Then make Episode 1 of a continuation that picks up after Part 3. Hook the viewer in the first two seconds and end on a cliffhanger. - Keep the characters exactly as they look in the references: same faces, hair, costumes and cinematic look. - Both characters are adults. Keep it rivalry with tension, nothing sexual. - Make new footage and new music. Don't reuse clips or the song from the references. - Dialogue and on-screen text in the same language as the references. Deliver the final video and the breakdown.LLMGrok 4.7Gen AIWan 3.0+6Skill—Softwarefal.ai5.6M tokens·26m 58s00
1 turnECEvan ChaOct 23Build a single-page web toy: a glossy black 3D pill button that says "let's collaborate" sits in the middle of a clean white studio page. The cursor is a cartoon white-glove hand. When the cursor touches the button and moves away, thick sticky black goo strands stretch from the button to the hand like melted tar or spider web, sag and wobble with real physics, thin out the further you pull, then snap and drip back into the button. Pressing and dragging should pull even more strands. Soft studio shadows, realistic glossy highlights on the button. One HTML file, no external assets. It should feel satisfying the moment it loads.LLMGrok 4.7Gen AI—Skill—Software—2.8M tokens·30m 51s00
Output C, 1 turnECEvan ChaOct 19Build a playable GTA-style 3D open-world mini city in the browser (Three.js + physics, single page, no external assets). Walk around in third person, steal and drive cars with arcade handling, traffic and pedestrians that react, a wanted-level system with police chases, a minimap and a day/night cycle. It should feel chaotic and fun from the first second.LLMGrok 4.7Gen AI—Skill—Software—4.6M tokens·40m 39s00
Output A, 1 turnECEvan ChaOct 19Build an interactive 3D browser sim of a Tesla Roadster made of jelly. Single HTML file; Three.js from a CDN is fine, no other external assets. Real soft-body feel: the body wobbles at speed, squishes on landings and jiggles when it hits walls. Drive with arrow keys or WASD, Space to jump, in a small arena with ramps and walls to bounce off. Glossy, translucent jelly look with good lighting. It should feel fun the moment it loads. · 4LLMGrok 4.7Gen AI—Skill—Software—12M tokens·52m 42s00
Output A, 1 turnECEvan ChaOct 8Build a playable 3D hover-bike racing game in the browser (Three.js). Neon cyberpunk city at night, rain, glowing billboards, bloom. Race through traffic with boost pads, ramps and near-miss sparks; speedometer, lap timer and a dramatic crash + replay screen. Make it look like a premium console game trailer the moment it loads. · 4LLMGrok 4.7Gen AI—Skill—Software—3M tokens·5m 13s00
Output C, 1 turnECEvan ChaOct 8Create a 12-second cinematic 3D product reveal film in the browser (Three.js) for a fictional smartwatch called ORBIT. Dark studio, dramatic rim lighting, slow camera orbit and dolly, liquid-metal reflections, a particle burst on the reveal, bold kinetic typography, logo end card. Auto-plays and loops like an Apple keynote launch film.LLMGrok 4.7Gen AI—Skill—Software—6M tokens·1h 14m 22s00
Output B, 1 turnECEvan ChaOct 8Build a playable 3D monster-truck smash game in the browser (Three.js + physics). Drive a giant monster truck through a colourful toy city and smash everything: towers of blocks topple, cars flip, glass shatters into debris with real physics, slow-motion on big crashes, dust and sparks, a destruction score. Should feel explosive and satisfying from the first second.LLMGrok 4.7Gen AI—Skill—Software—5.4M tokens·38m 23s00
Output C, 1 turnECEvan ChaOct 8Build a playable 3D flying game in the browser (Three.js): pilot a paper plane through floating islands at golden hour. Soft volumetric clouds, waterfalls spilling off the islands, flocks of birds, glowing rings to fly through, wind trails, a cinematic intro fly-by and a score. Hand-painted storybook look, gorgeous from the first frame.LLMGrok 4.7Gen AIGPT Image 2Skill—SoftwareChatGPT Images5.9M tokens·33m 30s00
Output A, 1 turnECEvan ChaOct 7Make a ~60s 16:9 product demo video for Evals (evals.ag): battle AI setups on your own task and vote blind. ## Story 1. Hook (first 3s must be instantly clear): real AI-made game/video outputs of the SAME prompt slam in side by side, moving, full-bleed → big type "Which one is better?" → rapid A/B/C/D flips on the beats → "Same prompt. Different setups." → "You can't tell by guessing." 2. A real game task is typed into the Evals composer: "Build a polished playable Sonic-style 3D platform game for the browser." → Battle → up to 4 setups (Model · Skill · Environment), each landing with real values and moving UI. 3. Send → velocity-matched zoom-through into the big hit → the 4 results play side by side. 4. Blind compare: two moving outputs, "Which would you choose?" → "I prefer A". 5. Run record of that same output: LLM, skill, environment, turn time, tokens. 6. Setup leaderboard: "Setups, not just models." 7. Community wall of real game / video / shorts outputs. 8. End card: Evals mark + wordmark, "Stop guessing. Battle it.", evals.ag. ## Content - Show only games, videos, 3D scenes, motion pieces and shorts. No websites, no landing pages. - Real Evals UI only, never mocked up. Every name and number on screen is real; only model names that exist in Evals (e.g. Claude Opus 5.5, GPT-6 Astra). - No real-person likeness. ## Voice - One energetic, confident English male narrator: excited on the hook, curious on the question, emphatic on the tagline, quick through the UI steps. - ~170 words in one continuous take that fills the whole film. No pause longer than ~0.35s, no clipped words. Say "Evals" clearly as "E-vals". - The edit follows the voice: each shot change lands on the beat nearest the start of the phrase it illustrates. ## Music & sound - Modern minimal tech-launch track, 120 BPM: punchy kick, deep 808, crisp hats, glassy synth plucks. Apple / Linear keynote feel. No cheesy EDM risers, no corporate stock vibe. - Voice always leads. Music dips ~6dB only while the voice speaks, on a smooth envelope. - SFX sit just under the music and rise above it only on hits. - 4–6 big hits get a layered reverse swell + sub boom + transient + glitch. Light ticks and whooshes elsewhere, never on every cut. ## Look - One sans font, one accent colour. - Real UI as rounded cards (16–22px radius) with soft shadows. - Sections alternate dark and light. - Never: outlines or hairline boxes, rules, underlines, highlight boxes behind words, cheap gradients, emoji, fake UI. ## Motion: this is the feel - Premium, fast, confident: an Apple / Linear keynote cut at shorts speed. - One continuous camera with velocity-continuous easing. Two easing personalities only: snappy (~25% of remaining distance per frame) and floaty (~12%). - Nothing ever stops dead. Every hold keeps drifting 1–3%. Fast moves decelerate over at least 6 frames. The end card keeps pushing until the last frame. - Exits accelerate into the cut; the next shot carries that velocity through it. - Real whip-pans: both shots travel together over 0.3–0.45s, motion blur peaking mid-move. One whip direction throughout. - Overlapping action: the next element starts while the previous one is ~60% done. No start-stop chains. - Text is screen-locked and arrives together with its container. Words reveal one by one, 2–4 frames apart, and the newest word flashes the accent. In ≤0.35s, readable ≥0.7s, out ≤0.25s. - On busy footage the camera stays calm (slow push ≤3%); the energy comes from cutting on the beat. - Every cut lands on the beat (±1 frame), with the visual hit and the SFX on the same frame. No double cuts. - Big hits (4–6 only): +8.5% zoom punch, ~16px shake at ~22Hz settling in 0.1s, RGB split 16px → 7px over 4 frames, 1–2 frames of white flash. A lighter version on secondary cuts. ## Quality bar Before calling it done, watch it like a motion director who didn't build it: - no dead stops, frozen frames or stutters - every cut on the beat - every word audible - every name and number real Fix the 5 worst moments first.LLMGrok 4.7Gen AIEleven v3+2Skill/hyperframesSoftwarefal.ai6.9M tokens·37m 4s00
Output A, 1 turnECEvan ChaOct 6Make a ~60s product launch film for Evals (evals.ag) — the place to battle AI setups on your own task and vote blind. 16:9, 1920×1080, 30fps, music-locked, no narration — on-screen type tells the story. Build it in HyperFrames (HTML + one paused GSAP timeline) under videos/evals-<name>/. 0. Inputs Product: https://evals.ag (app: https://evals-app.vercel.app — use my logged-in Chrome session for anything behind sign-in). Motion reference: videos/evals-hype/renders/evals-ref2-short-fast2-impact.mp4 (approved). Match its feel exactly; change only scenario, design and assets. Core message: "Same prompt, different setups — battle them and vote blind. Stop guessing. Battle it." What Evals really does (only show these — no invented features, numbers, rankings or names) Ask: type what you want to make in the composer ("What would you like to make?"). Run modes: Battle (blind), Side-by-side comparison (pick models and skills), Direct. Battle: up to 4 setups. A setup = model + skill + environment (e.g. Claude Opus 5.5 / GPT-6 Astra, skills like hyperframes / frontend-design / web-design-guidelines, environments like Native Claude / Recorded session). Compare, blind: same prompt, outputs side by side as playable live previews, names hidden until you vote ("I prefer A / B"). Inspect: every output keeps its full run record — LLM, gen AI, skill, software, environment, services contacted, turn time, turn tokens, status. Rank: votes roll up into leaderboards of setups, not just models ("Setups, not just models." — same model, different setups). Explore: community Outputs, Tasks, Skills, Compare, Community, Leaderboards. Real model names visible in the product: Claude Opus 5.5, GPT-6 Astra. Show only names/numbers you actually see in your captures. End card: Evals mark + wordmark, "Stop guessing. Battle it.", URL evals.ag. 1. Reverse-engineer the reference first (frames are the truth) Download the reference and study it frame by frame. Write SPEC.md with measured numbers, not impressions: every cut (frame number), shot order and durations, average shot length, cuts per 10s camera moves per shot (start/end position, scale, curve), easing measured per frame (fraction of remaining distance covered each frame) type-on speed (frames per char), caret behaviour, word-by-word reveal spacing, new-word accent flash duration card/UI choreography (fly-in paths, stacking, depth), light↔dark section switches, split-screen layouts sound map: where every hit/whoosh/click lands relative to the cut Then rebuild shot for shot at the same rhythm. Only brand, copy, design and assets change. 2. Scenario Write SCRIPT.md before building: a fresh narrative arc true to what Evals really does (hook: a new SOTA model every week — which one is actually best at your work? → type a task → Battle up to 4 setups → blind side-by-side, vote → full run record → setup leaderboards → "Stop guessing. Battle it." evals.ag). Don't reuse the previous films' copy. Short punchy lines (2–5 words per beat), English. Every beat gets a timestamp on the music grid. 3. Design Write DESIGN.md once and make every section obey it (no per-section drift): palette (one accent per frame), type pairing (one sans for headlines; weights/sizes in px; tracking −0.02em, line-height 1.0) card treatment: real UI captures, rounded 14–18px corners, soft layered shadow Never: outline/hairline boxes, rules, underlines, highlight boxes behind words, cheap gradients, emoji, fake UI mockups, invented numbers/rankings/features. 4. Assets (all real, all new) Capture the real Evals UI with Playwright at 2–3× DPR: home composer (empty / typed), Run-mode menu (Battle / Side-by-side / Direct), setups stepper (1–4), a live run page (4 agents generating → "Your options are ready"), Compare (blind A/B with "I prefer A/B"), output run record panel, Leaderboards (setup rows), Explore Outputs/Tasks/Skills. Run one real 4-setup Battle on a visual prompt (e.g. a playable browser game) so the four results are genuinely the same task. Record real Evals outputs as clips from their live previews (screencast, 1600×900, 30fps, 6–7s each; press start/arrow keys or orbit so they move). Use outputs not used in earlier Evals films. Typing scenes: put a text layer exactly over the real input field, matching its font, size and colour. New music track (detect BPM and first downbeat with a script) and a fresh SFX set. 5. Motion rules (this is the 감도 — non-negotiable) One camera = one pure function of time cam(t) built from keyframes with velocity-continuous easing; applied from the master timeline. Deterministic: no timers, no randomness, any frame renders alone. Two easing families only: snappy ≈25% of remaining distance per frame (90% in 8–9 frames) and floaty ≈12% (18–20 frames). No overshoot unless the reference has it. Exits accelerate into the cut; the next shot carries the motion through it (velocity-matched). Real whip-pans: outgoing and incoming shots travel together over 0.30–0.45s, with velocity-driven motion blur that peaks mid-move. Never blur switched on over a hard cut. Overlays are screen-locked (never inherit the camera whip). Text arrives with its container in one mask reveal. Overlapping action: the next element starts at ~60% of the previous tween. No start-stop chains. Nothing freezes: every hold carries a 1–3% slow drift (sine.inOut across the whole hold). Fast moves decelerate over ≥6 frames — no dead stops. No double cuts (two visual changes on adjacent frames). Busy footage gets only a slow push; energy comes from the cuts. Word reveals 2–4 frames apart; the newest word flashes the accent, then settles 3–12 frames later. Hero type-on 1 char / 2 frames; in-product typing 0.6–0.9 char/frame; caret solid (never blinks). Every cut on the beat ±1 frame; visual hit and SFX on the same frame. 6. Speed, edit, impact Speed by re-rendering, never frame-dropping: render at --fps 30/speedup and play back at 30fps (e.g. --fps 95/4 → 1.263×). Pick the speed-up so one beat = a whole number of frames. Quantize every segment to 8th notes; round segment start frames up. Impact FX in post (ffmpeg) on 4–6 big hits only (cut to dark, reveal word, flash, key word, logo): zoom punch +8.5%, decaying τ = 0.09s crop shake 16px at ~22Hz, decaying τ = 0.1s rgbashift 16px for 2 frames, then 7px for 2 frames 1–2 frames of white flash Lighter version on secondary cuts (3.5% punch, 6px shake, 1 frame of split). Find hit frames from frame-diff peaks and by looking. 7. Sound Music bed carries the film; no ducking. SFX bus ~3dB under the music overall, above it only on hits and in the music's own gaps. At each big hit, layer reverse swell + boom (high-passed at 60Hz) + short transient + glitch synced to the RGB split. Use a modest total (~40–45 sounds); never spread SFX evenly on every cut. Master at −14 LUFS, AAC 192k. 8. QA before you report (mandatory) motion.py <render> <bpm> <first_beat> must show: no dead stops, no frozen holds, every cut within ±1.5 frames of a beat, no adjacent-frame cuts. strip.sh (16 consecutive frames) at every transition, and look at each one. A good whip shows a smooth velocity ramp with both shots visible mid-move. Contact sheets alone hide motion bugs. Check text readability, safe margins, and that every number/name on screen exists in the real product. Deliver renders/evals-<name>-v1.mp4 plus a report: script, design summary, asset list, shot list with times, BPM and speed factor, hit frames with FX, SFX map, QA results.LLMGrok 4.7Gen AI—Skill/hyperframesSoftware—12.9M tokens·59m 42s00