Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when reviewing templates, rendered HTML, or shared components related to Remove comments and debug code in production. Validate the final browser-facing markup, not just the source framework abstraction.
.claude/skills/thedaviddias-clean-up-comments/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-03 | ✗→✓ | ▲ Improved | -31% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -12% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -32% | 0% |
| case-19 | ✗→✓ | ▲ Improved | -19% | 0% |
| case-02 | ✓→✓ | ＝ Same ✓ | -7% | 0% |
Comments expose internal logic to attackers, increase file size, and can leak sensitive information like API endpoints, credentials, or system architecture.
Review this HTML code for unnecessary comments, debug code, console logs, and temporary markup that should be removed before production deployment.
Remove all development comments, debug code, unused markup, and temporary elements from the HTML before deploying to production.
Explain why cleaning up code before production is important for security, performance, and professional presentation of web applications.
Review templates, server-rendered HTML, and shared components that output markup related to Remove comments and debug code in production. Flag exact elements, attributes, and routes where the rendered HTML violates the rule.
For full implementation details, code examples, and framework-specific guidance, see references/rule.md.
Rule page: https://frontendchecklist.io/en/rules/html/clean-up-comments
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-03 | fail→pass | 9,922 | 6,299 | -37% | 1 | 1 | 0% | 1,778 | 1,227 | -31% | 0 | 0 | — |
case-01 | fail→fail | 8,859 | 5,534 | -38% | 1 | 1 | 0% | 1,464 | 1,262 | -14% | 0 | 0 | — |
case-02 | pass→pass | 7,728 | 5,201 | -33% | 1 | 1 | 0% | 1,374 | 1,279 | -7% | 0 | 0 | — |
case-04 | fail→fail | 10,015 | 8,479 | -15% | 1 | 1 | 0% | 1,590 | 1,579 | -1% | 0 | 0 | — |
case-05 | pass→pass | 8,363 | 6,958 | -17% | 1 | 1 | 0% | 1,455 | 1,097 | -25% | 0 | 0 | — |
case-06 | fail→pass | 8,613 | 6,967 | -19% | 1 | 1 | 0% | 1,669 | 1,462 | -12% | 0 | 0 | — |
case-07 | fail→pass | 6,365 | 3,469 | -45% | 1 | 1 | 0% | 1,221 | 836 | -32% | 0 | 0 | — |
case-08 | pass→pass | 11,857 | 9,521 | -20% | 1 | 1 | 0% | 2,205 | 2,139 | -3% | 0 | 0 | — |
case-09 | fail→fail | 11,513 | 8,871 | -23% | 1 | 1 | 0% | 1,896 | 1,603 | -15% | 0 | 0 | — |
case-10 | pass→pass | 13,216 | 11,634 | -12% | 1 | 1 | 0% | 2,067 | 2,140 | +4% | 0 | 0 | — |
case-11 | pass→pass | 8,727 | 5,934 | -32% | 1 | 1 | 0% | 1,443 | 1,264 | -12% | 0 | 0 | — |
case-12 | pass→pass | 12,891 | 10,488 | -19% | 1 | 1 | 0% | 2,098 | 2,303 | +10% | 0 | 0 | — |
case-13 | pass→pass | 14,562 | 11,708 | -20% | 1 | 1 | 0% | 2,388 | 2,097 | -12% | 0 | 0 | — |
case-14 | fail→fail | 6,722 | 4,122 | -39% | 1 | 1 | 0% | 1,098 | 847 | -23% | 0 | 0 | — |
case-15 | fail→fail | 12,426 | 10,600 | -15% | 1 | 1 | 0% | 2,077 | 2,121 | +2% | 0 | 0 | — |
case-16 | pass→pass | 11,441 | 7,290 | -36% | 1 | 1 | 0% | 2,005 | 1,484 | -26% | 0 | 0 | — |
case-17 | pass→pass | 9,039 | 4,441 | -51% | 1 | 1 | 0% | 1,675 | 920 | -45% | 0 | 0 | — |
case-18 | pass→pass | 13,931 | 9,663 | -31% | 1 | 1 | 0% | 2,227 | 1,827 | -18% | 0 | 0 | — |
case-19 | fail→pass | 7,897 | 5,432 | -31% | 1 | 1 | 0% | 1,370 | 1,115 | -19% | 0 | 0 | — |
case-20 | fail→fail | 11,297 | 10,295 | -9% | 1 | 1 | 0% | 2,313 | 2,227 | -4% | 0 | 0 | — |
case-21 | pass→pass | 6,381 | 4,428 | -31% | 1 | 1 | 0% | 1,162 | 1,077 | -7% | 0 | 0 | — |
case-22 | pass→pass | 13,263 | 11,843 | -11% | 1 | 1 | 0% | 2,316 | 2,356 | +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. 22 cases were attempted. The headline lift of +18 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.