Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Generate and pressure-test headline options across proven formulas. Use when asked for headlines, a title, a subject line, a hook, or to improve a weak headline for a page, post, email, or ad. Produces 10–15 headline options grouped by formula (benefit, how-to, number, question, curiosity, social proof), each scored for clarity and specificity, with the top 3 recommended and why.
.claude/skills/mohitagw15856-headline-options/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-03 | ✗→✓ | ▲ Improved | -14% | 0% |
| case-01 | ✗→✓ | ▲ Improved | 10% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 6% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -5% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 30% | 0% |
The headline does 80% of the work — most people read it and decide. This skill generates a range of headlines across proven formulas (so you're not betting on one), scores them for clarity and specificity (the two things that actually drive clicks), and recommends the strongest — for a landing page, blog post, email subject, ad, or video title.
Ask for these only if they aren't already provided:
Options by formula (10–15 total), grouped:
Score each on Clarity and Specificity (1–5), since vague + clever loses to clear + specific:
| Headline | Formula | Clarity | Specific | Note | |---|---|---|---|---|
Top 3 picks — the strongest, with one line each on why, and which to A/B first.
If subject-line / limited — flag any that exceed the character budget.
Headline-writing practice (Ogilvy, Advertising's clarity-over-cleverness, the 4 U's) + formula-driven ideation and A/B discipline.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-03 | fail→pass | 15,181 | 10,478 | -31% | 1 | 1 | 0% | 3,015 | 2,583 | -14% | 0 | 0 | — |
case-04 | pass→pass | 14,574 | 15,791 | +8% | 1 | 1 | 0% | 2,486 | 3,125 | +26% | 0 | 0 | — |
case-05 | pass→fail | 11,351 | 16,739 | +47% | 1 | 1 | 0% | 2,073 | 3,681 | +78% | 0 | 0 | — |
case-01 | fail→pass | 14,629 | 14,297 | -2% | 1 | 1 | 0% | 2,931 | 3,214 | +10% | 0 | 0 | — |
case-02 | fail→pass | 14,212 | 9,861 | -31% | 1 | 1 | 0% | 2,628 | 2,775 | +6% | 0 | 0 | — |
case-06 | pass→fail | 5,839 | 11,070 | +90% | 1 | 1 | 0% | 1,401 | 2,802 | +100% | 0 | 0 | — |
case-07 | pass→pass | 14,231 | 11,552 | -19% | 1 | 1 | 0% | 3,030 | 2,961 | -2% | 0 | 0 | — |
case-08 | fail→pass | 13,026 | 9,984 | -23% | 1 | 1 | 0% | 2,643 | 2,506 | -5% | 0 | 0 | — |
case-09 | pass→pass | 13,779 | 10,561 | -23% | 1 | 1 | 0% | 2,867 | 2,748 | -4% | 0 | 0 | — |
case-10 | fail→pass | 11,070 | 11,020 | -0% | 1 | 1 | 0% | 2,150 | 2,794 | +30% | 0 | 0 | — |
case-11 | pass→pass | 15,590 | 17,323 | +11% | 1 | 1 | 0% | 3,450 | 4,660 | +35% | 0 | 0 | — |
case-12 | fail→pass | 14,822 | 11,130 | -25% | 1 | 1 | 0% | 2,742 | 2,756 | +1% | 0 | 0 | — |
case-13 | pass→pass | 12,372 | 11,206 | -9% | 1 | 1 | 0% | 2,786 | 2,968 | +7% | 0 | 0 | — |
case-14 | fail→fail | 14,282 | 14,732 | +3% | 1 | 1 | 0% | 2,866 | 3,626 | +27% | 0 | 0 | — |
case-15 | fail→fail | 13,979 | 12,235 | -12% | 1 | 1 | 0% | 2,505 | 3,051 | +22% | 0 | 0 | — |
case-16 | fail→pass | 13,733 | 9,274 | -32% | 1 | 1 | 0% | 2,715 | 2,433 | -10% | 0 | 0 | — |
case-17 | fail→pass | 11,121 | 12,404 | +12% | 1 | 1 | 0% | 2,115 | 2,911 | +38% | 0 | 0 | — |
case-18 | fail→pass | 12,560 | 12,120 | -4% | 1 | 1 | 0% | 2,731 | 3,097 | +13% | 0 | 0 | — |
case-19 | pass→pass | 13,347 | 12,598 | -6% | 1 | 1 | 0% | 2,693 | 3,228 | +20% | 0 | 0 | — |
case-20 | fail→pass | 13,457 | 11,703 | -13% | 1 | 1 | 0% | 2,475 | 2,731 | +10% | 0 | 0 | — |
case-21 | fail→pass | 14,328 | 11,486 | -20% | 1 | 1 | 0% | 2,984 | 2,936 | -2% | 0 | 0 | — |
case-22 | pass→pass | 13,076 | 13,831 | +6% | 1 | 1 | 0% | 2,168 | 3,116 | +44% | 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 +41 percentage points is the difference between those two pass rates over the 22 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.