Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when auditing slow page loads, heavy assets, or rendering delays related to Use secure and up-to-date JS libraries. Verify the actual bottleneck in DevTools, Lighthouse, or field data before recommending changes.
.claude/skills/thedaviddias-js-libraries/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-06 | ✗→✓ | ▲ Improved | 2% | 0% |
| case-20 | ✗→✓ | ▲ Improved | -5% | 0% |
| case-01 | ✓→✓ | ＝ Same ✓ | 4% | 0% |
| case-02 | ✓→✓ | ＝ Same ✓ | 2% | 0% |
| case-03 | ✓→✓ | ＝ Same ✓ | -9% | 0% |
Outdated or vulnerable libraries pose security risks and often include unnecessary bloat that degrades performance.
Check the project's JavaScript dependencies for known vulnerabilities and outdated versions.
Update vulnerable libraries to secure versions and replace bloated dependencies with modern, lightweight alternatives.
Explain the risks of using outdated JavaScript libraries and the benefits of using lightweight alternatives.
Review the routes, assets, and loading behavior that affect Use secure and up-to-date JS libraries. Flag exact files, requests, or rendering steps that add unnecessary network, CPU, or layout cost, and describe the measurement method used to confirm the issue.
For full implementation details, code examples, and framework-specific guidance, see references/rule.md.
Rule page: https://frontendchecklist.io/en/rules/performance/js-libraries
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 11,895 | 10,331 | -13% | 1 | 1 | 0% | 2,445 | 2,541 | +4% | 0 | 0 | — |
case-02 | pass→pass | 7,816 | 6,896 | -12% | 1 | 1 | 0% | 1,445 | 1,479 | +2% | 0 | 0 | — |
case-03 | pass→pass | 11,431 | 9,352 | -18% | 1 | 1 | 0% | 2,073 | 1,882 | -9% | 0 | 0 | — |
case-04 | pass→pass | 17,237 | 8,514 | -51% | 1 | 1 | 0% | 2,219 | 1,838 | -17% | 0 | 0 | — |
case-05 | pass→pass | 13,910 | 9,101 | -35% | 1 | 1 | 0% | 2,419 | 2,058 | -15% | 0 | 0 | — |
case-06 | fail→pass | 16,953 | 19,241 | +13% | 1 | 1 | 0% | 2,129 | 2,174 | +2% | 0 | 0 | — |
case-07 | pass→pass | 14,155 | 17,458 | +23% | 1 | 1 | 0% | 2,276 | 1,616 | -29% | 0 | 0 | — |
case-08 | pass→pass | 12,657 | 12,089 | -4% | 1 | 1 | 0% | 1,779 | 2,122 | +19% | 0 | 0 | — |
case-09 | pass→pass | 17,307 | 17,714 | +2% | 1 | 1 | 0% | 2,844 | 2,986 | +5% | 0 | 0 | — |
case-10 | pass→pass | 14,971 | 12,013 | -20% | 1 | 1 | 0% | 2,231 | 2,335 | +5% | 0 | 0 | — |
case-11 | pass→pass | 16,454 | 15,662 | -5% | 1 | 1 | 0% | 2,460 | 2,512 | +2% | 0 | 0 | — |
case-12 | pass→pass | 18,636 | 14,539 | -22% | 1 | 1 | 0% | 3,184 | 2,860 | -10% | 0 | 0 | — |
case-13 | pass→pass | 17,715 | 14,806 | -16% | 1 | 1 | 0% | 2,662 | 2,567 | -4% | 0 | 0 | — |
case-14 | pass→pass | 12,597 | 10,235 | -19% | 1 | 1 | 0% | 1,980 | 1,861 | -6% | 0 | 0 | — |
case-15 | pass→pass | 14,107 | 12,302 | -13% | 1 | 1 | 0% | 1,853 | 2,043 | +10% | 0 | 0 | — |
case-16 | pass→pass | 17,188 | 13,482 | -22% | 1 | 1 | 0% | 2,337 | 2,166 | -7% | 0 | 0 | — |
case-17 | pass→pass | 19,282 | 18,674 | -3% | 1 | 1 | 0% | 2,720 | 2,911 | +7% | 0 | 0 | — |
case-18 | pass→pass | 11,005 | 10,374 | -6% | 1 | 1 | 0% | 2,090 | 2,213 | +6% | 0 | 0 | — |
case-19 | pass→pass | 15,723 | 13,571 | -14% | 1 | 1 | 0% | 2,028 | 2,149 | +6% | 0 | 0 | — |
case-20 | fail→pass | 18,331 | 16,849 | -8% | 1 | 1 | 0% | 2,527 | 2,404 | -5% | 0 | 0 | — |
case-21 | pass→pass | 25,382 | 24,268 | -4% | 1 | 1 | 0% | 3,367 | 3,686 | +9% | 0 | 0 | — |
case-22 | pass→pass | 18,327 | 13,155 | -28% | 1 | 1 | 0% | 2,691 | 2,184 | -19% | 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 +9 percentage points is the difference between those two pass rates over the 22 comparable cases.
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.