Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Готовит многомерный разбор конкурентов — feature-матрица, SWOT с cross-strategy, 5 сил Портера, сравнение ценообразования, позиционирование, прогноз стратегических ходов и три уровня дифференциации (догнать / отстроиться / создать новое). Адаптируется под цель (продуктовый дизайн / fundraising / стратегия / годовой обзор). User-invoked only — do NOT auto-trigger. Triggers on /pm-competitive, "конкурентный анализ", "разбор конкурентов", "five forces", "SWOT", "competitive analysis", "competitor c
.claude/skills/serejaris-pm-competitive/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-03 | ✗→✓ | ▲ Improved | -9% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 222% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 77% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 96% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 108% | 0% |
Part of the Personal Corp framework — running a one-person business through AI agents. Run a systematic multi-dimensional competitor study. Branches by analysis purpose. Includes information-credibility tagging, time-stamping of all findings, and explicit separation of fact vs inference.
| Field | Required | Notes | |---|---|---| | Competitor list | yes | 2-5 names; > 5 → batch into rounds | | Our product | no | Used for differentiation positioning | | Purpose | no | Product design / fundraising deck / strategic planning / annual review; default product design | | Dimensions | no | Features / pricing / UX / tech / business model; default features + pricing |
| Purpose | Emphasis | Output focus | Length | |---|---|---|---| | Product design | Features + UX | Feature matrix + differentiation list | 2-3 pages | | Fundraising deck | Market structure + moats | Landscape map + moat analysis | 1 page | | Strategic planning | Full depth | SWOT + 5-Forces + pricing + roadmap | 5-8 pages | | Annual review | Market share change + trends | YoY changes + trend judgment | 2-3 pages |
Use user-supplied materials first (URLs, screenshots, pricing pages). Tag everything with credibility and timestamp.
Credibility tiers:
| Tier | Sources | Tagging | |---|---|---| | High | Official site, pricing page, official announcements | Quote directly | | Medium | Media coverage, user reviews, industry reports | Cite source | | Low | Rumors, speculation, outdated info | Mark [unverified] |
Information-collection checklist:
| Dimension | What to collect | Typical sources | |---|---|---| | Fundamentals | Founded, funding rounds, team size, user base | Official site, business databases | | Product | Core feature list, recent updates, roadmap | Official site, changelog, blog | | Pricing | Tiers, price ranges, free-tier limits | Pricing page | | Reputation | Praise, complaints, NPS | Review sites, app stores, social platforms | | Strategy | Target market, acquisition channels, partnerships | Press, social media |
Always tag freshness — e.g. "based on Q1 2026 public information" — and prompt user to verify currency.
Evaluate from industry-structure angle.
| Force | Dimension | Evaluation points | |---|---|---| | Supplier power | Upstream dependency | Concentration of key tech/talent/resources; switching cost | | Buyer power | Downstream customer leverage | Customer concentration; switching cost; price sensitivity | | New-entrant threat | Entry barriers | Tech / capital / brand / network effects / regulation | | Substitute threat | Alternative solutions | What's the user's current alternative? Substitute's value-for-money? | | Industry rivalry | Existing-player dynamics | Competitor count, concentration, differentiation, exit barriers |
Output: each force = strong / medium / weak + one-sentence rationale. Overall industry attractiveness = high / medium / low.
Feature comparison matrix (core deliverable):
| Module | Sub-feature | Us | Comp A | Comp B | Comp C | |---|---|---|---|---|---| | {module 1} | {sub 1} | ✅ Mature | ✅ Mature | ⚠️ Basic | ❌ None | | | {sub 2} | ⚠️ Basic | ✅ Mature | ✅ Mature | ⚠️ Basic |
Legend: ✅ Mature / ⚠️ Basic (exists but limited) / ❌ Missing / 🔜 Planned
Positioning deep-dive:
| Dimension | Method | Scoring | |---|---|---| | Feature score | 1-5 per module, weighted total | 5 = best-in-class, 3 = passing, 1 = barely present | | UX comparison | Steps in core flow, learning curve, completion efficiency | Compare clicks + time-to-complete | | Tech architecture | Patterns, stack, performance, openness | API openness, integration capability, extensibility |
SWOT per competitor:
| | Positive | Negative | |---|---|---| | Internal | Strengths (core advantages) | Weaknesses (clear gaps) | | External | Opportunities (exploitable) | Threats (must watch) |
Each quadrant: 2-3 items, each with factual support. Banned: vague claims like "strong team" or "advanced tech".
SWOT cross-strategies:
| Strategy | Meaning | Direction | |---|---|---| | SO | Strengths × Opportunities | Use strengths to grab market window | | WO | Weaknesses × Opportunities | Fill gaps to capture new opportunities | | ST | Strengths × Threats | Use moats to defend against attacks | | WT | Weaknesses × Threats | Most urgent defensive priorities |
Pricing comparison (if pricing dimension included):
| Item | Us | A | B | C | |---|---|---|---|---| | Free-tier scope | {} | {} | {} | {} | | Entry-tier monthly | {} | {} | {} | {} | | Enterprise monthly | {} | {} | {} | {} | | Billing model | {per-seat / usage / feature-tier} | | | | | Core differentiator | {pricing strategy reading} | | | |
Pricing-strategy types:
Pick two most-discriminating dimensions (e.g. "feature depth vs ease of use", or "price vs feature breadth") and place all products into a 2×2:
| Quadrant | Trait | Products | |---|---|---| | High features + high price | Enterprise full-stack | {list} | | High features + low price | Best value | {list} | | Low features + high price | Niche specialist | {list} | | Low features + low price | Entry-level | {list} |
Predict competitor's likely next moves based on recent signals.
Method:
| Competitor | Recent moves | Inferred intent | Impact on us | Confidence | Recommended response | |---|---|---|---|---|---| | {A} | {actions} | {hypothesis} | {assessment} | High/Medium/Low | {strategy} |
Projection rules:
| Tier | Definition | Action | |---|---|---| | Catch-up | Competitors have it; we don't | List features to add + priority + estimated effort | | Differentiate | We're unique or clearly ahead | List strengths to reinforce + marketing message ideas | | Innovate | No competitor has done it; opportunity exists | List blue-ocean candidates + validation approach |
[unverified]| Structure | Trait | Strategy advice | |---|---|---| | One-dominant | One player > 50% share | Differentiate niches, avoid head-on | | Duopoly | Top 2 = > 70% combined | Pick a side, or be the third option | | Fragmented | Top 5 each < 20% | Speed-grab a niche #1 position | | Emerging | No clear leader | Educate market first, build brand |
| Frequency | What to monitor | Trigger | |---|---|---| | Weekly | Product changelog, social signals | Major updates → tracker entry | | Monthly | Pricing changes, new features, press | Update feature matrix | | Quarterly | Full competitor report, strategy projection refresh | Quarterly competitor brief | | Event-driven | Funding / M&A / major launches / exec changes | Immediate impact assessment |
/pm-prd — differentiation gap → PRD/pm-prioritize — catch-up feature list → RICE-rank/pm-feedback — user mentions of competitors → enrich perception data/pm-brainstorm — diverge on differentiation ideas| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-21 | pass→pass | 19,992 | 28,011 | +40% | 1 | 1 | 0% | 4,019 | 7,792 | +94% | 0 | 0 | — |
case-01 | fail→fail | 54,986 | 43,235 | -21% | 1 | 1 | 0% | 8,300 | 8,735 | +5% | 0 | 0 | — |
case-02 | fail→fail | 60,459 | 47,584 | -21% | 1 | 1 | 0% | 8,278 | 9,327 | +13% | 0 | 0 | — |
case-03 | fail→pass | 48,466 | 30,495 | -37% | 1 | 1 | 0% | 8,260 | 7,504 | -9% | 0 | 0 | — |
case-04 | fail→pass | 25,635 | 25,365 | -1% | 1 | 1 | 0% | 1,578 | 5,077 | +222% | 0 | 0 | — |
case-05 | fail→pass | 27,057 | 34,221 | +26% | 1 | 1 | 0% | 3,740 | 6,606 | +77% | 0 | 0 | — |
case-06 | fail→pass | 26,962 | 38,736 | +44% | 1 | 1 | 0% | 4,322 | 8,468 | +96% | 0 | 0 | — |
case-07 | fail→pass | 23,863 | 26,575 | +11% | 1 | 1 | 0% | 3,216 | 6,694 | +108% | 0 | 0 | — |
case-08 | fail→pass | 23,316 | 27,612 | +18% | 1 | 1 | 0% | 3,153 | 6,477 | +105% | 0 | 0 | — |
case-09 | fail→fail | 23,766 | 37,503 | +58% | 1 | 1 | 0% | 3,138 | 7,557 | +141% | 0 | 0 | — |
case-10 | fail→fail | 22,413 | 32,566 | +45% | 1 | 1 | 0% | 3,402 | 7,428 | +118% | 0 | 0 | — |
case-11 | pass→pass | 17,941 | 28,857 | +61% | 1 | 1 | 0% | 2,853 | 6,246 | +119% | 0 | 0 | — |
case-12 | fail→pass | 27,051 | 20,195 | -25% | 1 | 1 | 0% | 2,523 | 5,491 | +118% | 0 | 0 | — |
case-13 | fail→pass | 20,293 | 30,403 | +50% | 1 | 1 | 0% | 3,158 | 7,334 | +132% | 0 | 0 | — |
case-14 | fail→pass | 16,138 | 29,053 | +80% | 1 | 1 | 0% | 3,094 | 7,042 | +128% | 0 | 0 | — |
case-15 | fail→pass | 31,700 | 26,862 | -15% | 1 | 1 | 0% | 4,960 | 6,464 | +30% | 0 | 0 | — |
case-16 | fail→pass | 17,098 | 26,267 | +54% | 1 | 1 | 0% | 2,628 | 6,426 | +145% | 0 | 0 | — |
case-17 | fail→pass | 44,518 | 36,295 | -18% | 1 | 1 | 0% | 2,375 | 7,983 | +236% | 0 | 0 | — |
case-18 | fail→fail | 21,717 | 37,183 | +71% | 1 | 1 | 0% | 3,207 | 8,174 | +155% | 0 | 0 | — |
case-19 | fail→fail | 49,349 | 39,430 | -20% | 1 | 1 | 0% | 3,171 | 7,325 | +131% | 0 | 0 | — |
case-20 | fail→pass | 36,290 | 35,117 | -3% | 1 | 1 | 0% | 4,634 | 7,837 | +69% | 0 | 0 | — |
case-22 | pass→pass | 38,839 | 37,144 | -4% | 1 | 1 | 0% | 6,953 | 7,450 | +7% | 0 | 0 | — |
case-23 | fail→fail | 2,684 | 8,430 | +214% | 1 | 1 | 0% | 385 | 2,995 | +678% | 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 +57 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.