Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Apply a UX lens to a user-facing change: whether it serves the user's real goal and whether the path through it holds together, using the Understanding, Bridging, and Flowing contexts. Use when scoping, planning, or assessing any change that affects what a user sees or does. Loaded as a lens during planning and assessment.
.claude/skills/tobihagemann-user-experience/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | 45% | 0% |
| case-16 | ✗→✓ | ▲ Improved | 6% | 0% |
| case-15 | ✓→✗ | ▼ Worse | 16% | 0% |
| case-22 | ✓→✗ | ▼ Worse | 21% | 0% |
| case-12 | ✓→✓ | ＝ Same ✓ | 13% | 0% |
Apply this lens to any user-facing change, both before committing to an approach and when judging a built experience. Keep the work anchored to what the user is trying to accomplish.
Lead with the user's goal rather than a list of surface fixes. A pile of small corrections ("relabel this", "move that") is a sign the underlying flow was never examined. Start from intent and work down to detail.
Scope: this lens covers whether a change serves the user and whether the path through it is coherent. Visual and aesthetic craft — typography, color, spacing, motion polish — is a separate concern and stays out of scope here.
Establish what the user actually needs before shaping a solution.
Translate the need into the right solution rather than the first one that comes to mind.
Trace the path the user takes from end to end.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-12 | pass→pass | 12,957 | 13,598 | +5% | 1 | 1 | 0% | 2,205 | 2,493 | +13% | 0 | 0 | — |
case-01 | pass→pass | 19,689 | 21,072 | +7% | 1 | 1 | 0% | 3,766 | 3,931 | +4% | 0 | 0 | — |
case-02 | pass→pass | 17,022 | 16,080 | -6% | 1 | 1 | 0% | 3,330 | 3,481 | +5% | 0 | 0 | — |
case-03 | pass→pass | 14,244 | 14,081 | -1% | 1 | 1 | 0% | 2,721 | 2,700 | -1% | 0 | 0 | — |
case-04 | fail→pass | 11,503 | 14,620 | +27% | 1 | 1 | 0% | 1,821 | 2,642 | +45% | 0 | 0 | — |
case-05 | pass→pass | 15,027 | 14,925 | -1% | 1 | 1 | 0% | 2,308 | 2,626 | +14% | 0 | 0 | — |
case-06 | pass→pass | 14,183 | 13,937 | -2% | 1 | 1 | 0% | 2,149 | 2,380 | +11% | 0 | 0 | — |
case-07 | pass→pass | 15,144 | 13,561 | -10% | 1 | 1 | 0% | 2,283 | 2,340 | +2% | 0 | 0 | — |
case-08 | fail→fail | 14,078 | 11,185 | -21% | 1 | 1 | 0% | 2,098 | 2,091 | -0% | 0 | 0 | — |
case-09 | fail→fail | 13,645 | 13,990 | +3% | 1 | 1 | 0% | 2,016 | 2,450 | +22% | 0 | 0 | — |
case-10 | fail→fail | 3,913 | 5,751 | +47% | 1 | 1 | 0% | 661 | 1,238 | +87% | 0 | 0 | — |
case-11 | pass→pass | 13,365 | 15,287 | +14% | 1 | 1 | 0% | 1,966 | 2,651 | +35% | 0 | 0 | — |
case-13 | pass→pass | 12,557 | 13,026 | +4% | 1 | 1 | 0% | 1,870 | 2,493 | +33% | 0 | 0 | — |
case-14 | pass→pass | 13,898 | 15,250 | +10% | 1 | 1 | 0% | 2,073 | 2,644 | +28% | 0 | 0 | — |
case-15 | pass→fail | 14,642 | 20,516 | +40% | 1 | 1 | 0% | 2,149 | 2,498 | +16% | 0 | 0 | — |
case-16 | fail→pass | 14,555 | 12,603 | -13% | 1 | 1 | 0% | 1,987 | 2,116 | +6% | 0 | 0 | — |
case-17 | pass→pass | 14,993 | 13,902 | -7% | 1 | 1 | 0% | 2,248 | 2,399 | +7% | 0 | 0 | — |
case-18 | pass→pass | 11,059 | 13,599 | +23% | 1 | 1 | 0% | 1,808 | 2,358 | +30% | 0 | 0 | — |
case-19 | pass→pass | 15,421 | 13,969 | -9% | 1 | 1 | 0% | 2,294 | 2,515 | +10% | 0 | 0 | — |
case-20 | pass→pass | 13,001 | 10,892 | -16% | 1 | 1 | 0% | 2,013 | 2,002 | -1% | 0 | 0 | — |
case-21 | pass→pass | 15,031 | 14,024 | -7% | 1 | 1 | 0% | 2,470 | 2,442 | -1% | 0 | 0 | — |
case-22 | pass→fail | 13,122 | 14,159 | +8% | 1 | 1 | 0% | 2,043 | 2,465 | +21% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of 0 percentage points is the difference between those two pass rates over the 22 comparable cases. 2 cases got worse with the skill loaded, and they are included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.