Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Record a learned design constraint from a correction into the project brief. Invoke when the user asks for remember on their UI, or mentions 'remember' alongside design / UI / frontend work.
.claude/skills/educlopez-remember/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-08 | ✗→✓ | ▲ Improved | 600% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 306% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 60% | 0% |
| case-17 | ✗→✓ | ▲ Improved | 769% | 0% |
| case-18 | ✗→✓ | ▲ Improved | 1030% | 0% |
<!-- HARNESS MIRROR — do not edit here. Canonical source: skills/ or commands/. After editing source, copy into cli/assets/<harness>/ and repo-root harness mirrors. -->
Context: this sub-skill is one lens of the broader ui-craft skill. If the ui-craft skill is also installed, read its SKILL.md first for Discovery + Anti-Slop + Craft Test, then apply the specific lens below.
Load references/brief.md (sections 6 + Self-Correction) before proceeding.
From the argument (or the immediately preceding correction in the conversation), capture:
Append a dated entry to section 6 (Learned constraints) of .ui-craft/brief.md:
markdown- **YYYY-MM-DD** — <the rule>. *Why:* <reason>.
Create the brief (run /brief) or the section 6 heading if absent. If the new constraint contradicts an existing principle or constraint, mark the old one deprecated with a dated reason (brief is append-mostly) rather than deleting it. Never store a rule that breaches the accessibility/correctness floor — record the closest compliant interpretation and say so.
If the user signals this should apply to all their projects ("in all my projects", "whenever you work with me"), that's general memory, outside the brief's project scope:
Report in one line where it landed and what it changes, e.g.: > Recorded in the brief (learned constraint): never gradients on a hero here.
Next step: Keep working. The constraint now lives in the brief, so every future session reads it (rung 2).
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-06 | fail→fail | 5,592 | 8,196 | +47% | 1 | 1 | 0% | 250 | 998 | +299% | 0 | 0 | — |
case-11 | fail→fail | 3,050 | 7,168 | +135% | 1 | 1 | 0% | 446 | 961 | +115% | 0 | 0 | — |
case-01 | fail→fail | 2,406 | 37,890 | +1475% | 1 | 1 | 0% | 387 | 1,118 | +189% | 0 | 0 | — |
case-02 | fail→fail | 5,323 | 8,351 | +57% | 1 | 1 | 0% | 767 | 939 | +22% | 0 | 0 | — |
case-03 | fail→fail | 2,639 | 7,258 | +175% | 1 | 1 | 0% | 432 | 1,098 | +154% | 0 | 0 | — |
case-04 | fail→fail | 6,041 | 9,203 | +52% | 1 | 1 | 0% | 969 | 1,160 | +20% | 0 | 0 | — |
case-05 | fail→fail | 4,940 | 28,976 | +487% | 1 | 1 | 0% | 237 | 1,026 | +333% | 0 | 0 | — |
case-07 | pass→pass | 10,676 | 7,494 | -30% | 1 | 1 | 0% | 1,678 | 1,612 | -4% | 0 | 0 | — |
case-08 | fail→pass | 2,314 | 7,136 | +208% | 1 | 1 | 0% | 231 | 1,617 | +600% | 0 | 0 | — |
case-09 | fail→pass | 2,535 | 5,184 | +104% | 1 | 1 | 0% | 302 | 1,226 | +306% | 0 | 0 | — |
case-10 | fail→fail | 2,836 | 3,943 | +39% | 1 | 1 | 0% | 284 | 1,103 | +288% | 0 | 0 | — |
case-12 | fail→fail | 3,091 | 6,008 | +94% | 1 | 1 | 0% | 360 | 916 | +154% | 0 | 0 | — |
case-13 | fail→pass | 6,153 | 5,517 | -10% | 1 | 1 | 0% | 861 | 1,375 | +60% | 0 | 0 | — |
case-14 | pass→fail | 8,193 | 11,314 | +38% | 1 | 1 | 0% | 1,353 | 1,169 | -14% | 0 | 0 | — |
case-15 | fail→fail | 4,148 | 5,028 | +21% | 1 | 1 | 0% | 491 | 1,226 | +150% | 0 | 0 | — |
case-16 | fail→fail | 10,820 | 8,680 | -20% | 1 | 1 | 0% | 1,327 | 923 | -30% | 0 | 0 | — |
case-17 | fail→pass | 1,709 | 9,419 | +451% | 1 | 1 | 0% | 155 | 1,347 | +769% | 0 | 0 | — |
case-18 | fail→pass | 5,241 | 12,241 | +134% | 1 | 1 | 0% | 244 | 2,757 | +1030% | 0 | 0 | — |
case-19 | fail→fail | 4,772 | 8,373 | +75% | 1 | 1 | 0% | 207 | 1,043 | +404% | 0 | 0 | — |
case-20 | pass→pass | 6,695 | 6,595 | -1% | 1 | 1 | 0% | 1,141 | 1,584 | +39% | 0 | 0 | — |
case-21 | pass→pass | 10,062 | 9,218 | -8% | 1 | 1 | 0% | 1,840 | 1,816 | -1% | 0 | 0 | — |
case-22 | pass→pass | 11,407 | 8,804 | -23% | 1 | 1 | 0% | 1,854 | 2,108 | +14% | 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, and 11 counted toward the lift figure. The other 11 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +18 percentage points is the difference between those two pass rates over the 11 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.