Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →code-saurabh/design-explorer
Visual design exploration and design system creation skill.
Passed 18 of 23 test cases with the skill, and 8 of 23 without it.
and 15 more in the full report
11 fixed7 passed either way4 still failing1 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 11 test cases the skill fixed. Same model, same prompt, with and without the skill.
I'm building a mobile web view for a personal finance app aimed at young adults who want to start investing. I haven't settled on a visual direction yet and want to see several distinct design ideas before deciding. Please generate a range of initial concept options for me. For each direction, give it a descriptive name reflecting its visual personality, specify exact CSS font stacks and type hierarchy, detail a complete hex color scheme with role assignments, outline the spacing approach and key layout decisions, highlight context-specific risks, and rate each across major usability and design criteria. Finish with a side-by-side comparison grid across all directions and prompt me on how to combine elements I like.
| Checked | Without skill | With skill |
|---|---|---|
| Generates between 4 and 6 distinct design variants | Missed | Passed |
| Includes a mandatory contrarian variant exploring an unexpected or anti-obvious direction | Missed | Passed |
| Includes a comparison table rating variants across distinctiveness, legibility, emotional resonance, scalability, technical feasibility, and overall score | Missed | Passed |
The instructions your model reads when the skill loads.
A full-spectrum visual design partner. Combines the rapid variant generation of /design-shotgun, the systematic quality of /design-html, and the consultative rigour of /design-consultation into one unified, opinionated workflow.
Invoke proactively when the user:
Before producing a single variant, collect enough signal to avoid generic output.
Ask (or infer from context) the following — do NOT skip any item:
If the session context already contains prior taste memory or approved variants, load that data instead of asking again.
test cases passed, vs 8 / 23 without
Published skills whose descriptions closely match this one, with their headline benchmark lift.
More skills from the same source and nearby in the registry, with their headline benchmark lift.