Install any skill in seconds. Free to start, no credit card required.
Get Started Free →velcrafting/spec-sculptor
name: spec-sculptor
Passed 16 of 21 test cases with the skill, and 4 of 21 without it.
and 18 more in the full report
16 fixed1 passed either way2 still failing3 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 15 test cases the skill fixed. Same model, same prompt, with and without the skill.
I have a dump of ideas for a public wishlist sharing system: users should save items for later, generate a unique shareable URL for friends, and receive price-drop alerts. However, we shouldn't support collaborative editing or private password-protected lists in this version. Also, we need to consider what happens if a saved product gets discontinued. Could you organize these notes into a structured feature document outlining the problem context, target scenarios, goals, non-goals, functional details, data entities, potential edge cases, and risks with mitigations?
| Checked | Without skill | With skill |
|---|---|---|
| The response excludes numbered section prefixes on top-level section headings. | Missed | Passed |
| The response formats risks alongside paired mitigations rather than inside a table. | Missed | Passed |
| The response contains zero programming code or SQL DDL statements. | Passed | Passed |
The instructions your model reads when the skill loads.
name: spec-sculptor description: Convert a storm of thoughts into a concise feature spec suitable for planning and implementation. metadata: short-description: Ideas → spec mode: spec read-only: true
Produce a concise spec that prevents implementation-by-guessing.
<What problem are we solving and why now?>
test cases passed, vs 4 / 21 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.