AI Model Reviewer

GPT-6 Sol in Codex CLI 0.156.1 + Remotion/HyperFrames skills

skills on · Codex CLI 0.156.1 + Remotion/HyperFrames skillsKit: hyperframes fc9071c1fc79 (Apache-2.0) · 3 of 21 installed skills used in the session

← Open in the gallerySame prompt, all models

Rendered video (by the agent)

Poster for rendered video (by the agent)

Rendered video (by the agent). This is the finished MP4 saved in the run. Played from a 720p web copy made by this site from the agent's out.mp4; the original file is too large for this host and is kept unchanged on the runs branch. Download video (1.4 MB).

OKGlass dashboard launchmax effortNeutral harness

Wall time 23m 33sCost $3.23 estimated

Review

No verdict yet.

The full review

OK. Finished on its own after 23.6 min.

Scores

AI judge

Not scored

No rubric for this prompt

Visitor votes and reviewer grades are not available yet.

AI judge
AI judge: the test runner (model claude-opus-5-5), scoring by visual review of the 1600 px render against the pre-registered rubric. No AI-judge score exists for this result: there is no rubric for this prompt, so no score is shown and none is made up.
Visitors
Public voting with a blind reveal is coming: you score the output first, then see the other scores.
Reviewer
The reviewer's own grade appears here once it is added.

Scores compare results for the same prompt only. AI-judge boards rank runs within one prompt, equal totals sharing a rank; scores from different prompts are never averaged.

Downloads

Files actually published for this run. Original output, site-made previews and post-run conversions are labelled separately.

The exact prompt

as run, 3,447 characters. Every result for this prompt

[Task]
Create a 24-second product-launch video for a fictional analytics app called Lumora, at 1920x1080, 30 fps. A frosted-glass dashboard assembles panel by panel, comes alive with data and ends on a logo lockup.

[Scenes]
1. 0-3 s: Darkness. Three soft colour orbs drift into place and bloom. A thin glass sheet slides up from below, catching a moving specular highlight. The text "Your numbers, finally clear." fades in word by word.
2. 3-8 s: The camera pulls back and tilts slightly as the dashboard builds. A sidebar, a header, four KPI cards and two large chart panels drop in with staggered springs, each with a blurred backdrop and a 1 px light border.
3. 8-13 s: The charts animate. The revenue line draws left to right with a gradient fill rising under it, the channel donut sweeps into segments, and the KPI numbers count up from zero.
4. 13-18 s: A cursor glides in on a curved path. It hovers over the September point and a tooltip springs out, then it clicks the "Region" tab. The bar chart morphs bars into their new heights, and a toggle flips to dark-glass mode.
5. 18-21 s: The camera dollies in on the retention card, where the circular progress ring fills to 92%. All the panels then fan apart in 3D depth.
6. 21-24 s: The panels fold into a single glass tile that becomes the Lumora logomark (a stylised lens of three overlapping arcs). The wordmark and the tagline "Analytics you can see through" settle in.

[Style]
Palette: #0B0F1F night, #7C5CFF violet, #22D3EE cyan, #F472B6 rose, #F8FAFC frost white. Clean geometric sans with tabular figures for every number. Glass panels use backdrop blur, a subtle noise layer, inner highlights and soft coloured shadows. Motion uses spring entries with a light overshoot, a 60 ms stagger between siblings and ease-in-out camera moves.

[Data]
Monthly revenue ($k), Apr-Sep: 42, 51, 48, 63, 71, 88.
KPIs: Active users 18,420 · MRR $88.2k · Churn 1.8% · NPS 64.
Channels: Organic 41%, Referral 27%, Paid 19%, Partners 13%.
Regions (signups): North 3,120 · South 2,480 · East 4,050 · West 2,760.
Retention: 92%.

[Wow]
- Real parallax depth: the background orbs, the panels and the cursor move at three distinct depths during camera moves.
- A specular light sweep travels across every glass edge in sync with each panel landing.
- The bar chart morph interpolates both height and colour, with no pop or jump.
- The panel-to-logo fold is one continuous, shape-preserving transformation.

[Rules]
- Every frame is drawn in code (SVG, canvas, CSS or WebGL). No image files, no image generation, no external assets beyond fonts.
- Deterministic: no unseeded randomness at render time.
- Text must be crisp, spelled correctly and readable at 1080p.
- No hard cuts between scenes: every change is a designed transition.
- All charts must use the exact values given in the Data section.

[Harness notes] Render the final video to out.mp4 (1920x1080, 24 s). Write everything into the working directory. HyperFrames 0.8.70 is already installed for you in the current directory: node_modules is pre-installed and read-only, so do not run npm install or npx --yes; use the local hyperframes command (for example hyperframes init <dir> --example blank, then hyperframes render). Render the final video yourself to out.mp4 in the current directory. You may take snapshots and inspect your own output (headless chromium is available at /usr/bin/chromium) and iterate on your own before finishing.

How it was made

Wall time 23m 33sFirst artifact at 22m 19s1 saved version

Versions saved during the run

1 intermediate output, in the order the model wrote them.

Version 1 of 1
  1. Version 1 at 22m 21s thumbnailv1 22:21

Step by step

The prompt, every agent turn and tool call (tool name and a one-line summary; inputs and outputs are not shown), then the final message. The full transcript is 44.0 MB; it is published as a scrubbed, compressed download rather than shown on the page.

Show all 178 steps

Loading the timeline…

Download the full transcript Codex CLI rollout JSONL, 589 KB, gzip. Scrubbed: anything that looks like a key, token, e-mail address or phone number is replaced, and account details and the test machine's file paths are removed.

Cost breakdown

Headline cost $3.23, estimated from the token counts at list price.

Tokens by typeTokens$ per 1MCost
Input (uncached)12,850$2.00$0.0257
Cache write472,589$2.00$0.9452
Cache read6,819,621$0.20$1.3639
Output89,723$10.00$0.8972
of which reasoning50,752billed as output
Total computed from tokens
$3.2320
Total reported by the harness
none

Computed from harness-reported token counts x list prices. Cache writes priced at the input rate (the price table has no separate cache-write price). No request above 272,000 input tokens (77 checked): standard rates.

Prices: Project price table, 2026-09-25.

Run notes

one prompt, agentic harness (self-review allowed)

  • frozen prompt (written): written 2026-10-03 for MR wave rh (owner ask m:d1f2d829 + m:1122fe24: more Remotion and HyperFrames, ambitious multi-scene, harness runs); concept 'glass_dashboard_launch' (hyperframes only); all frames code-drawn; standard hyperframes harness suffix (24 s); zero results
  • plain CODEX_HOME copy (config.toml and framework skills, run work/ trusted; no AGENTS.md/state)
  • W2 hyperframes skills installed in this run's codex_home/skills
  • W2: the framework's pinned node_modules is bound read-only at work/node_modules (the prompt's 'pre-installed and read-only'); Node 22.23.3 and work/node_modules/.bin first on PATH; headless Chromium /usr/bin/chromium
  • Mode: interactive TUI in tmux (160x48), inside bubblewrap sandbox
  • Isolation audit: outside paths seen