AI Model Reviewer

Claude Opus 5.5 in OpenCode 1.18.32

skills on · OpenCode 1.18.32 + default skills kitKit: default (46 web/Three.js/QA + 33 video) · 79 skills installed; skill use not recorded

← Open in the gallerySame prompt, all models

Poster frame of Claude Opus 5.5'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.

Stopped by usSaas growth metricsmax effortNeutral harness

Wall time 42m 41sCost $9.45 reported

Review

No verdict yet.

The full review

Stopped by us. Stopped by us (box row cap reached (2850s); stopped by the box lane so the run ships as partial); not a model failure.

Scores

Not scored: only runs that finished on their own are scored.

Downloads

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

The exact prompt

as run, 3,757 characters. Every result for this prompt

Build an interactive SaaS growth dashboard from the exact dataset below (CSV: one row per month and plan with customers at the start of the month, new and churned customers during it, and the monthly list price per customer in US dollars). Embed the data in the page verbatim and compute every number from it; do not invent or alter any value. Ending customers = customers_start + new_customers - churned_customers, and MRR = ending customers x price_usd.

month,plan,customers_start,new_customers,churned_customers,price_usd
2025-05,Starter,220,36,15,19
2025-05,Team,90,10,3,79
2025-05,Enterprise,12,3,0,499
2025-06,Starter,241,39,15,19
2025-06,Team,97,12,3,79
2025-06,Enterprise,15,1,1,499
2025-07,Starter,265,46,16,19
2025-07,Team,106,9,4,79
2025-07,Enterprise,15,1,0,499
2025-08,Starter,295,33,24,19
2025-08,Team,111,18,3,79
2025-08,Enterprise,16,2,0,499
2025-09,Starter,304,29,18,19
2025-09,Team,126,13,4,79
2025-09,Enterprise,18,3,0,499
2025-10,Starter,315,47,23,19
2025-10,Team,135,16,6,79
2025-10,Enterprise,21,2,1,499
2025-11,Starter,339,36,20,19
2025-11,Team,145,13,4,79
2025-11,Enterprise,22,3,0,499
2025-12,Starter,355,45,25,19
2025-12,Team,154,17,6,79
2025-12,Enterprise,25,0,0,499
2026-01,Starter,375,38,26,19
2026-01,Team,165,16,4,79
2026-01,Enterprise,25,2,0,499
2026-02,Starter,387,28,31,19
2026-02,Team,177,15,25,79
2026-02,Enterprise,27,1,0,499
2026-03,Starter,384,31,25,19
2026-03,Team,167,17,5,79
2026-03,Enterprise,28,2,1,499
2026-04,Starter,390,29,20,19
2026-04,Team,179,9,6,89
2026-04,Enterprise,29,1,0,549
2026-05,Starter,399,32,34,19
2026-05,Team,182,16,6,89
2026-05,Enterprise,30,2,0,549
2026-06,Starter,397,30,33,19
2026-06,Team,192,12,8,89
2026-06,Enterprise,32,0,1,549
2026-07,Starter,394,43,27,19
2026-07,Team,196,15,8,89
2026-07,Enterprise,31,0,1,549
2026-08,Starter,410,41,28,19
2026-08,Team,203,17,7,89
2026-08,Enterprise,30,0,0,549
2026-09,Starter,423,43,35,19
2026-09,Team,213,15,5,89
2026-09,Enterprise,30,2,0,549
2026-10,Starter,431,28,25,19
2026-10,Team,223,19,7,89
2026-10,Enterprise,32,3,1,549

The dashboard needs: KPI cards for current MRR, MRR growth over the period, current customers, average monthly logo churn rate and net revenue retention estimate; a stacked area chart of MRR by plan over time; a waterfall chart that decomposes the most recent month's MRR change into new, churned and price-change effects; a churn-rate line per plan with any month above twice its plan's average flagged and explained in an annotation; a cohort-style table of customers by plan and month with heat colouring; plan and date-range filters that update every chart; hover or tap tooltips with exact values; and a short auto-written insights panel (three sentences) computed from the data, such as which plan drives growth and when prices changed. Use a refined, modern product-analytics look. So the page is alive without input, on load the charts animate in and then a guided tour cycles forever: it highlights each KPI, steps a cursor through the months on the area chart with live tooltips, rebuilds the waterfall, applies and clears a plan filter, and points out the churn anomaly, until the user touches a control.

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 42m 41sFirst artifact at 34m 43s1 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 34m 50s thumbnailv1 34:50

No transcript was delivered with this result.

Cost breakdown

Headline cost $9.45, as reported by OpenCode (its own bill for the run). The list-price estimate from the token counts is $9.45.

Tokens by typeTokens$ per 1MCost
Input (uncached)64$4.00$0.0003
Cache write622,702$5.00$3.1135
Cache read7,112,525$0.20$1.4225
Output245,549$20.00$4.9110
of which reasoning183,242billed as output
Total computed from tokens
$9.4473
Total reported by the harness
$9.4473
Difference
matches

OpenCode 1.18.32 reported its own total; token counts x list prices shown beside it. The harness reports reasoning tokens separately from output; here output includes them (62,307 visible + 183,242 reasoning), and both bill at the output rate.

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 data_dashboards; zero results
  • plain OpenCode config copy (providers only; no ECC instructions/agents)
  • default skills on: 79 skills in this run's xdg/config/opencode/skills
  • opencode export written to a file (not piped)
  • Mode: interactive TUI in tmux (160x48), inside bubblewrap sandbox
  • Isolation audit: none
  • Renderer: inline SVG (every chart is hand-built SVG markup written by the page's own script); no chart library, no canvas, nothing loaded from a CDN (single self-contained file).
  • Card video: the run's own video.mp4 (H.264 yuv420p, unchanged). Poster: the run's own render.png.
  • Status: stopped by us. Stopped at the row cap (~50 min) before the harness declared it finished; the page shown is what it had built by then.
  • 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.