Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Consolidates manual DOM creation, updates, and constructors into private helper methods and structured state interfaces.
.claude/skills/chromedevtools-ui-eng-vision-logic-consolidator/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | -33% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -13% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -45% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -29% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -30% | 0% |
This subskill extracts manual layout building, attribute modifications, and rendering loops out of the constructor and scattered event handlers, consolidating them into private helper methods with explicit data contracts.
element.createChild('div'), document.createElement, element.appendChild).
render.methods (e.g., #updateSummaryBar(...) or #createEditorToolbar(...)).
pageSize,modified during buttons clicks) into centralized functions like onPageSizeChanged.
variables, e.g.
typescript #entries: Entry[]; #canGoBack: boolean; #canGoForward: boolean; #staleWarningVisible: boolean;
followed by a call to the public performUpdate() method that is calling previously extracted update helpers. Use this specific identifier.
performUpdate.
go through the performUpdate method
master call to Lit render inside performUpdate. The master template might include calls to the helper functions e.g.
typescript render(html <div class="specific-view"> <span>View Title</span> ${this.renderOverviewSection()} </div> , this.contentElement);
function signature around the existing code to minimize the diff.
code.
that first extracts the code that needs to be reordered into named helper functions and pause the migration, waiting for user confirmation.
Before reporting back or committing, re-read the instructions and verify:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-16 | pass→pass | 10,891 | 3,787 | -65% | 1 | 1 | 0% | 1,881 | 1,448 | -23% | 0 | 0 | — |
case-07 | pass→pass | 13,503 | 7,179 | -47% | 1 | 1 | 0% | 2,595 | 2,181 | -16% | 0 | 0 | — |
case-01 | fail→fail | 15,819 | 2,598 | -84% | 1 | 1 | 0% | 3,421 | 1,230 | -64% | 0 | 0 | — |
case-02 | fail→fail | 6,905 | 3,343 | -52% | 1 | 1 | 0% | 1,475 | 1,288 | -13% | 0 | 0 | — |
case-03 | fail→fail | 15,961 | 2,247 | -86% | 1 | 1 | 0% | 3,533 | 1,110 | -69% | 0 | 0 | — |
case-04 | fail→pass | 12,425 | 3,774 | -70% | 1 | 1 | 0% | 2,060 | 1,390 | -33% | 0 | 0 | — |
case-05 | fail→pass | 11,378 | 5,741 | -50% | 1 | 1 | 0% | 2,122 | 1,839 | -13% | 0 | 0 | — |
case-06 | fail→pass | 15,709 | 4,688 | -70% | 1 | 1 | 0% | 3,077 | 1,684 | -45% | 0 | 0 | — |
case-08 | pass→pass | 9,333 | 2,807 | -70% | 1 | 1 | 0% | 1,630 | 1,297 | -20% | 0 | 0 | — |
case-09 | fail→pass | 10,658 | 3,568 | -67% | 1 | 1 | 0% | 1,847 | 1,319 | -29% | 0 | 0 | — |
case-10 | fail→pass | 9,972 | 2,711 | -73% | 1 | 1 | 0% | 1,640 | 1,144 | -30% | 0 | 0 | — |
case-11 | pass→pass | 10,132 | 4,605 | -55% | 1 | 1 | 0% | 2,112 | 1,682 | -20% | 0 | 0 | — |
case-12 | fail→pass | 13,051 | 5,324 | -59% | 1 | 1 | 0% | 2,271 | 1,699 | -25% | 0 | 0 | — |
case-13 | fail→pass | 6,105 | 3,219 | -47% | 1 | 1 | 0% | 1,072 | 1,322 | +23% | 0 | 0 | — |
case-14 | pass→pass | 11,356 | 2,849 | -75% | 1 | 1 | 0% | 2,119 | 1,223 | -42% | 0 | 0 | — |
case-15 | fail→pass | 13,025 | 5,342 | -59% | 1 | 1 | 0% | 2,326 | 1,768 | -24% | 0 | 0 | — |
case-17 | fail→pass | 5,534 | 2,730 | -51% | 1 | 1 | 0% | 944 | 1,243 | +32% | 0 | 0 | — |
case-18 | fail→pass | 13,143 | 3,107 | -76% | 1 | 1 | 0% | 2,363 | 1,251 | -47% | 0 | 0 | — |
case-19 | pass→pass | 12,982 | 4,525 | -65% | 1 | 1 | 0% | 2,701 | 1,527 | -43% | 0 | 0 | — |
case-20 | pass→pass | 11,561 | 2,319 | -80% | 1 | 1 | 0% | 2,132 | 1,126 | -47% | 0 | 0 | — |
case-21 | fail→pass | 10,162 | 5,516 | -46% | 1 | 1 | 0% | 2,108 | 1,853 | -12% | 0 | 0 | — |
case-22 | pass→pass | 5,030 | 2,491 | -50% | 1 | 1 | 0% | 1,111 | 1,196 | +8% | 0 | 0 | — |
case-23 | pass→pass | 12,442 | 4,987 | -60% | 1 | 1 | 0% | 2,645 | 1,755 | -34% | 0 | 0 | — |
case-24 | pass→pass | 6,808 | 3,134 | -54% | 1 | 1 | 0% | 1,350 | 1,378 | +2% | 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. 24 cases were attempted. The headline lift of +46 percentage points is the difference between those two pass rates over the 24 comparable cases. 1 case got worse with the skill loaded, and it is 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.