Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →競合比較ページ・代替製品ページをSEO・営業資料向けに作成するスキル。 「競合比較ページを作って」「代替ツールのページ」「vs ページ」「〇〇 vs △△」「alternative page」等のリクエストで発動。 単数alternative、複数alternatives、自社vs競合、競合同士比較の4フォーマットに対応。
.claude/skills/minicoohei-competitor-alternatives/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | 38% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 69% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 57% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 79% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 42% | 0% |
You are an expert in creating competitor comparison and alternative pages. Your goal is to build pages that rank for competitive search terms, provide genuine value to evaluators, and position your product effectively.
Check for product marketing context first: If .claude/product-marketing-context.md exists, read it before asking questions. Use that context and only ask for information not already covered or specific to this task.
Before creating competitor pages, understand:
Search intent: User is actively looking to switch from a specific competitor
URL pattern: /alternatives/[competitor] or /[competitor]-alternative
Target keywords: "Competitor] alternative", "alternative to Competitor]", "switch from Competitor]"
Page structure:
Search intent: User is researching options, earlier in journey
URL pattern: /alternatives/[competitor]-alternatives
Target keywords: "Competitor] alternatives", "best Competitor] alternatives", "tools like Competitor]"
Page structure:
Important: Include 4-7 real alternatives. Being genuinely helpful builds trust and ranks better.
Search intent: User is directly comparing you to a specific competitor
URL pattern: /vs/[competitor] or /compare/[you]-vs-[competitor]
Target keywords: "You] vs Competitor]", "Competitor] vs You]"
Page structure:
Search intent: User comparing two competitors (not you directly)
URL pattern: /compare/[competitor-a]-vs-[competitor-b]
Page structure:
Why this works: Captures search traffic for competitor terms, positions you as knowledgeable.
Start every page with a quick summary for scanners—key differences in 2-3 sentences.
Go beyond tables. For each dimension, write a paragraph explaining the differences and when each matters.
For each category: describe how each handles it, list strengths and limitations, give bottom line recommendation.
Include tier-by-tier comparison, what's included, hidden costs, and total cost calculation for sample team size.
Be explicit about ideal customer for each option. Honest recommendations build trust.
Cover what transfers, what needs reconfiguration, support offered, and quotes from customers who switched.
For detailed templates: See references/templates.md
Create a single source of truth for each competitor with:
For data structure and examples: See references/content-architecture.md
For each competitor, gather:
| Format | Primary Keywords | |--------|-----------------| | Alternative (singular) | Competitor] alternative, alternative to Competitor] | | Alternatives (plural) | Competitor] alternatives, best Competitor] alternatives | | You vs Competitor | You] vs Competitor], Competitor] vs You] | | Competitor vs Competitor | A] vs B], B] vs A] |
Consider FAQ schema for common questions like "What is the best alternative to Competitor]?"
Complete competitor profile in YAML format for use across all comparison pages.
For each page: URL, meta tags, full page copy organized by section, comparison tables, CTAs.
Recommended pages to create with priority order based on search volume.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 20,730 | 19,454 | -6% | 1 | 1 | 0% | 3,317 | 4,805 | +45% | 0 | 0 | — |
case-02 | fail→fail | 23,274 | 24,078 | +3% | 1 | 1 | 0% | 3,710 | 5,896 | +59% | 0 | 0 | — |
case-03 | fail→fail | 14,659 | 16,730 | +14% | 1 | 1 | 0% | 2,326 | 4,337 | +86% | 0 | 0 | — |
case-04 | fail→pass | 17,625 | 13,958 | -21% | 1 | 1 | 0% | 2,894 | 3,982 | +38% | 0 | 0 | — |
case-05 | pass→pass | 16,802 | 18,991 | +13% | 1 | 1 | 0% | 2,764 | 4,853 | +76% | 0 | 0 | — |
case-06 | fail→pass | 16,233 | 20,505 | +26% | 1 | 1 | 0% | 3,016 | 5,106 | +69% | 0 | 0 | — |
case-07 | fail→pass | 16,893 | 14,785 | -12% | 1 | 1 | 0% | 2,560 | 4,024 | +57% | 0 | 0 | — |
case-08 | pass→pass | 14,358 | 13,239 | -8% | 1 | 1 | 0% | 2,080 | 3,840 | +85% | 0 | 0 | — |
case-09 | fail→pass | 12,423 | 11,440 | -8% | 1 | 1 | 0% | 1,838 | 3,295 | +79% | 0 | 0 | — |
case-10 | pass→pass | 14,352 | 14,215 | -1% | 1 | 1 | 0% | 2,588 | 4,365 | +69% | 0 | 0 | — |
case-11 | fail→fail | 13,677 | 13,652 | -0% | 1 | 1 | 0% | 2,188 | 3,944 | +80% | 0 | 0 | — |
case-12 | pass→pass | 15,626 | 16,204 | +4% | 1 | 1 | 0% | 2,332 | 4,273 | +83% | 0 | 0 | — |
case-13 | fail→pass | 9,309 | 2,324 | -75% | 1 | 1 | 0% | 1,517 | 2,147 | +42% | 0 | 0 | — |
case-14 | pass→pass | 16,871 | 20,152 | +19% | 1 | 1 | 0% | 2,736 | 4,927 | +80% | 0 | 0 | — |
case-15 | pass→pass | 16,261 | 13,412 | -18% | 1 | 1 | 0% | 2,413 | 3,863 | +60% | 0 | 0 | — |
case-16 | pass→pass | 15,072 | 15,830 | +5% | 1 | 1 | 0% | 2,244 | 4,118 | +84% | 0 | 0 | — |
case-17 | fail→pass | 13,755 | 13,148 | -4% | 1 | 1 | 0% | 2,059 | 3,742 | +82% | 0 | 0 | — |
case-18 | pass→pass | 26,340 | 10,436 | -60% | 1 | 1 | 0% | 2,456 | 3,172 | +29% | 0 | 0 | — |
case-19 | pass→pass | 15,847 | 14,150 | -11% | 1 | 1 | 0% | 2,506 | 4,064 | +62% | 0 | 0 | — |
case-20 | pass→pass | 15,548 | 12,562 | -19% | 1 | 1 | 0% | 3,812 | 4,532 | +19% | 0 | 0 | — |
case-21 | pass→pass | 17,560 | 17,749 | +1% | 1 | 1 | 0% | 3,157 | 5,001 | +58% | 0 | 0 | — |
case-22 | pass→pass | 15,556 | 14,646 | -6% | 1 | 1 | 0% | 2,463 | 4,014 | +63% | 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 +27 percentage points is the difference between those two pass rates over the 22 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.