Verified against Claude · 2026-07-14
Turn a recurring small decision into a reusable Artifact widget
A compact prompt for building a small, single-purpose interactive Artifact — a checklist, scorer, or picker — for a decision you make repeatedly, so you get a reusable tool instead of a one-off chat answer.
The prompt
Ready to copy — highlighted parts are example details you can swap.
Build a small, single-purpose interactive Artifact, not a general chat answer. The whole point is that I can reuse this the next ten times I face this decision without re-explaining it. DECISION THIS HELPS WITH Whether to accept a freelance project based on rate, timeline, and how well it fits my niche. INPUTS THE TOOL SHOULD ASK FOR Offered rate, estimated hours, days until deadline, fit with niche (1-5 slider) HOW IT SHOULD DECIDE OR SCORE Effective hourly rate = offered rate / estimated hours. Score = effective hourly rate weighted 50%, fit weighted 30%, timeline slack weighted 20%. Below a combined score of 6/10, recommend declining. WHAT IT SHOULD OUTPUT A single recommendation label (Accept / Consider / Decline) plus the computed effective hourly rate shown plainly. REQUIREMENTS - Keep it to one screen, no scrolling required for the core interaction, unless the input list genuinely needs more room. - Every input should be a simple control — text field, number, slider, toggle, or select — nothing that requires instructions to operate. - The output should update as inputs change, so I can try a few scenarios quickly rather than re-running it each time. - Do not add features I did not ask for, even good ones — this should stay small enough that I can understand and trust every part of it.
Customize the highlighted detailsoptional — the prompt above already works
Why this works
This differs from a general 'build a tool' Artifact prompt by front-loading reusability as the actual success criterion, which changes what Claude optimizes for: a one-off answer can afford ambiguity because you will re-explain context next time, but a tool meant to survive ten future uses cannot, so naming that constraint up front pushes toward simple, self-explanatory controls over a clever one-time interface. Capping it at one screen and plain controls (text, number, slider, toggle, select) is a scope-control mechanic borrowed from constraint-based coding prompts generally, but it matters more here than in a one-off Artifact because a decision-support tool you'll open repeatedly needs to be operable from muscle memory, not re-read each time. The explicit 'do not add features I did not ask for, even good ones' line targets a specific Artifact failure mode: given room to build something interactive, models often add polish — extra fields, a history log, a chart — that increases surface area without being asked, and for a tool meant to be trusted and reused quickly, an unrequested feature is a liability, not a bonus, because it is one more thing that could be silently wrong.
Verified against
Claude Sonnet 4.6 (Artifacts) · 2026-07-14
Changelog
- 2026-07-14 — Initial publish, verified against Claude Sonnet 4.6 Artifacts.
Building this for real?
This is a free starting point. If you'd rather have what Scult builds built and running for your business, that's Scult's day job.
EXPLORE WHAT SCULT BUILDS
