AI Model Reviewer

GPT-6 Astra in Codex CLI 0.156.1

skills off

← Open in the gallerySame prompt, all models

Zoom, inline player and more ways to view
Poster frame of GPT-6 Astra's page
Open the original page

The model's own page, run in a sandboxed frame only when you ask (a page that needs its own origin opens in a new tab on a separate demo site). Phones get a full-screen player capped at 1.5x pixel density (2x on larger screens) so it stays smooth; the original page runs uncapped. Tap inside for sound.

OKMaze generate and solvemax effortNeutral harness

Wall time 16m 34sCost $3.95 estimated

Review

No verdict yet.

The full review

OK. Finished on its own after 16.6 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, 1,941 characters. Every result for this prompt

Build a maze generation and solving show: four panels side by side (a grid on narrow screens) each carve the same-size 31 x 21 maze with a different generator at the same time: recursive backtracker (depth-first), randomized Prim's, randomized Kruskal's with union-find, and Wilson's loop-erased random walk. Show each generator's working state honestly: the backtracker's stack as a glowing trail, Prim's frontier cells, Kruskal's sets as distinct colours that merge as walls fall, Wilson's wandering walk with its loops being erased. Count steps per panel. When a maze is complete, solve it in the same panel with A* from the top-left to the bottom-right corner, drawing the open set, the closed set and the final path as a bright line, and show path length and cells explored. Then display a small comparison (generation steps, dead ends, longest corridor, solution length) so viewers see how the generators produce different textures of maze. Controls: maze size presets, speed slider, play, pause, single step, regenerate, and a toggle to solve with breadth-first search instead of A*. So the page is alive without input, on load it generates and solves automatically, then waits 3 seconds and starts a new round with a fresh seed, forever, until the user touches a control. Make it feel like a calm, hypnotic museum installation.

Save it as index.html in the current directory: one self-contained file with inline CSS and JavaScript only, no libraries or frameworks, and no external requests of any kind (no CDNs, web fonts, images or audio files). It must work when opened directly from disk, fit any window from a 390 px wide phone to a desktop, and run without console errors. You may serve it locally and open it in headless chromium (/usr/bin/chromium, software WebGL, no GPU) to screenshot and inspect your own output, and iterate on your own before finishing. Close any browser you open as soon as you've taken your screenshot.

How it was made

Wall time 16m 34sFirst artifact at 8m 25s4 saved versions

Versions saved during the run

4 intermediate outputs, in the order the model wrote them.

Version 4 of 4
  1. Version 1 at 8m 26s thumbnailv1 8:26
  2. Version 2 at 11m 18s thumbnailv2 11:18
  3. Version 3 at 14m 15s thumbnailv3 14:15
  4. Version 4 at 15m 56s thumbnailv4 15:56

No transcript was delivered with this result.

Cost breakdown

Headline cost $3.95, estimated from the token counts at list price.

Tokens by typeTokens$ per 1MCost
Input (uncached)45$10.00$0.0004
Cache write74,774$10.00$0.7477
Cache read650,294$1.00$0.6503
Output51,067$50.00$2.5534
of which reasoning22,698billed as output
Total computed from tokens
$3.9518
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).

Prices: Project price table (vendor list prices), 2026-09-22.

Run notes

one prompt, agentic harness (self-review allowed)

  • frozen prompt (written): written 2026-10-05 for the MR category expansion (owner request: 500 finished runs in every category); category algorithm_visualizers; zero results
  • plain CODEX_HOME copy (config.toml only, run work/ trusted; no AGENTS.md/skills/state)
  • Mode: interactive TUI in tmux (160x48), inside bubblewrap sandbox
  • Isolation audit: outside paths seen
  • Renderer: Canvas 2D (getContext('2d') at index.html line 753) plus DOM/CSS UI and inline SVG; single self-contained index.html, no CDN and no external assets.
  • Card video: the run's own session capture (headless Chromium, no GPU), re-encoded as H.264. Poster: the run's own render.png.
  • Same frozen prompt as the other rows on this test; configurations differ and all rows are unscored, so this is not a head-to-head result.