Concept render — composed live from this prompt's palette & layout. Reproducible with the library, not a stock screenshot.
Concept scene: Star Motes — open in library →Agent run viewer with step rail
vibe: technical, transparent, calm
starter — 4 blocks, 1 stackdesktop viewport≈ 330 lines generated (estimate)≈ 600 tokens to run (estimate)
prompt.txt — copy-paste ready
Paste into:
This is the text every run below was produced from. The other targets wrap it for a specific tool — same brief, different framing.
An agent run viewer: a run header (task, model, total time), a vertical step rail (plan → act → verify) with per-step status and elapsed time, tool call cards under the act step showing tool name, input one-liner and a collapsible result, and a result pane. Timings are labelled as demo figures. Steps announce their status changes; focus is visible; colours are CSS variable tokens; reduced-motion collapses the stagger so steps appear without motion.
Palette tokens:#0a0c11#dbe3ec#22d3ee#a78bfa
Remix this brief:
- — Timeline: steps as a horizontal trace.
- — Minimal: rail only, no result pane.
- — Debug-flavoured: raw JSON view toggle.
Components this brief typically renders:
Community:▶ Re-run this prompt⤴ Submit a remix of it★ Your saved listRe-runs are a labelled simulation; the run log on this page is the real record.
Local preview
Not a model run. A deterministic render of this brief's own stated blocks, palette and viewport — so you can feel the brief before copying it.
⚠ Before you run it, know this
- Lowest score was Codex at 88/100 — Result pane honest about demo data.
- All models built cleanly on the first pass.
More prompts in this mood
Open prompt →
Multi-model comparison page
honest, tabular, technical
VerifiedReactNext.js
89/100 · 3 runsOpen prompt →
AI model status page
technical, transparent, calm
VerifiedHTMLReact
89/100 · 3 runsOpen prompt →
AI code review with inline diff
technical, precise, constructive
FeaturedReactNext.js
91/100 · 3 runs