Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Draft a persuasive public comment on a proposed rule, regulation, or plan. Use when asked to comment on a rulemaking, respond to a consultation, submit feedback on a proposed regulation, or write a comment to an agency. Produces a structured comment: your position, specific evidence-based arguments tied to the proposal's text, suggested edits, and the impact — the kind agencies must consider on the record.
.claude/skills/mohitagw15856-public-comment/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | 226% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 69% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 141% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 155% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 29% | 0% |
Agencies must review and respond to substantive comments — but only substantive ones move the needle. A comment that cites the specific provision, brings evidence, and proposes concrete alternative language carries far more weight than "I support/oppose this." This skill drafts that substantive comment.
Ask for these only if they aren't already provided:
Re / docket line — the proposal and docket/reference number, and your position in one line.
Who I am & my interest — brief; establishes standing and why your input is relevant.
Summary of position — what you support/oppose/want changed, up front.
Substantive comments — the core. Each point:
Number them so the agency can respond point by point.
Impact — the concrete effect (cost, burden, benefit, unintended consequence) on you/your community — this is what agencies weigh.
Conclusion & request — restate the specific action requested; offer to provide more info.
Notice-and-comment rulemaking practice (substantive, provision-specific, evidence-based comments with proposed alternatives).
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 42,103 | 24,352 | -42% | 1 | 1 | 0% | 5,734 | 4,146 | -28% | 0 | 0 | — |
case-02 | fail→pass | 8,642 | 23,986 | +178% | 1 | 1 | 0% | 1,019 | 3,319 | +226% | 0 | 0 | — |
case-03 | pass→pass | 19,456 | 16,508 | -15% | 1 | 1 | 0% | 2,078 | 2,731 | +31% | 0 | 0 | — |
case-04 | fail→pass | 23,742 | 29,691 | +25% | 1 | 1 | 0% | 2,246 | 3,790 | +69% | 0 | 0 | — |
case-05 | fail→fail | 14,540 | 19,626 | +35% | 1 | 1 | 0% | 1,123 | 2,801 | +149% | 0 | 0 | — |
case-06 | fail→pass | 13,910 | 25,310 | +82% | 1 | 1 | 0% | 1,341 | 3,229 | +141% | 0 | 0 | — |
case-07 | fail→pass | 14,616 | 21,234 | +45% | 1 | 1 | 0% | 1,347 | 3,438 | +155% | 0 | 0 | — |
case-08 | fail→pass | 18,705 | 21,229 | +13% | 1 | 1 | 0% | 2,151 | 2,783 | +29% | 0 | 0 | — |
case-09 | fail→pass | 115,451 | 31,163 | -73% | 1 | 1 | 0% | 2,310 | 4,190 | +81% | 0 | 0 | — |
case-10 | fail→fail | 14,109 | 19,011 | +35% | 1 | 1 | 0% | 1,192 | 2,654 | +123% | 0 | 0 | — |
case-11 | fail→pass | 25,072 | 32,768 | +31% | 1 | 1 | 0% | 3,150 | 4,083 | +30% | 0 | 0 | — |
case-12 | fail→pass | 17,620 | 24,464 | +39% | 1 | 1 | 0% | 2,422 | 3,211 | +33% | 0 | 0 | — |
case-13 | fail→fail | 26,581 | 25,092 | -6% | 1 | 1 | 0% | 3,001 | 3,779 | +26% | 0 | 0 | — |
case-14 | fail→fail | 27,067 | 62,608 | +131% | 1 | 1 | 0% | 3,093 | 4,456 | +44% | 0 | 0 | — |
case-15 | fail→pass | 19,376 | 25,434 | +31% | 1 | 1 | 0% | 2,496 | 4,104 | +64% | 0 | 0 | — |
case-16 | fail→pass | 17,382 | 23,308 | +34% | 1 | 1 | 0% | 2,672 | 3,802 | +42% | 0 | 0 | — |
case-17 | fail→fail | 14,599 | 19,806 | +36% | 1 | 1 | 0% | 2,560 | 3,188 | +25% | 0 | 0 | — |
case-18 | fail→fail | 19,126 | 18,697 | -2% | 1 | 1 | 0% | 2,797 | 3,116 | +11% | 0 | 0 | — |
case-19 | pass→pass | 16,826 | 16,999 | +1% | 1 | 1 | 0% | 2,323 | 3,111 | +34% | 0 | 0 | — |
case-20 | fail→pass | 19,486 | 19,570 | +0% | 1 | 1 | 0% | 2,592 | 3,654 | +41% | 0 | 0 | — |
case-21 | pass→pass | 18,370 | 19,754 | +8% | 1 | 1 | 0% | 3,014 | 3,989 | +32% | 0 | 0 | — |
case-22 | pass→pass | 29,836 | 26,409 | -11% | 1 | 1 | 0% | 3,996 | 5,481 | +37% | 0 | 0 | — |
case-23 | pass→pass | 23,433 | 30,882 | +32% | 1 | 1 | 0% | 3,002 | 3,605 | +20% | 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. 23 cases were attempted. The headline lift of +48 percentage points is the difference between those two pass rates over the 23 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.