Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Conduct SWOT analysis with TOWS matrix for strategic planning. Use this skill when the user needs to evaluate a company, product, or project by identifying strengths, weaknesses, opportunities, and threats, or wants to generate strategic options from internal and external factors. Also use when the user mentions competitive positioning, strategic assessment, or asks 'what are our advantages and risks', even without naming SWOT explicitly.
.claude/skills/asgard-ai-platform-biz-swot/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 21% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 212% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 117% | 0% |
| case-15 | ✗→✓ | ▲ Improved | 6% | 0% |
| case-19 | ✗→✓ | ▲ Improved | 107% | 0% |
SWOT structures strategic assessment by separating internal factors (Strengths, Weaknesses) from external factors (Opportunities, Threats). The TOWS matrix then crosses these four quadrants to generate concrete strategy options. This skill guides the full workflow: from factor identification through prioritization to actionable strategy formulation.
Trigger conditions:
When NOT to use:
IRON LAW: Internal vs External Classification
Strengths and Weaknesses are INTERNAL — they originate from within the
organization and the organization can directly control or change them.
Examples: proprietary technology, brand reputation, team expertise, cost
structure, operational processes.
Opportunities and Threats are EXTERNAL — they originate from the environment
and the organization cannot directly control them.
Examples: market trends, regulatory changes, competitor actions, economic
conditions, technological shifts.
NEVER classify an external trend as a Strength or Weakness.
NEVER classify an internal capability as an Opportunity or Threat.
Test: "Can the organization decide to change this tomorrow?"
YES → Internal (S or W)
NO → External (O or T)IRON LAW: Completeness Before Synthesis
Complete ALL four quadrants before generating strategies.
Do NOT jump to recommendations after identifying only strengths and
opportunities. Skipping W or T produces dangerously optimistic strategies.Examine the organization's resources, capabilities, and performance:
For each factor, classify as Strength (competitive advantage) or Weakness (competitive disadvantage) relative to competitors. Be specific — "good team" is too vague; "engineering team with 5+ years experience in ML deployment" is actionable.
Scan the environment across these dimensions:
For each factor, classify as Opportunity (favorable) or Threat (unfavorable). Ground each factor in evidence — cite data, trends, or specific events when possible.
Not all factors carry equal weight. For each quadrant, rank factors by:
Keep the top 3-5 factors per quadrant for the TOWS matrix. Including too many dilutes the analysis.
Cross the four quadrants to generate four types of strategies:
| | Opportunities | Threats | |---|---|---| | Strengths | SO strategies: Use strengths to capture opportunities | ST strategies: Use strengths to defend against threats | | Weaknesses | WO strategies: Address weaknesses to capture opportunities | WT strategies: Minimize weaknesses to avoid threats |
For each cell, generate 1-3 specific strategy options. Each strategy must reference at least one factor from each axis (e.g., SO strategy must name a specific strength AND a specific opportunity).
Convert the most promising TOWS strategies into actionable recommendations:
markdown# SWOT Analysis: {Subject} ## Internal Factors ### Strengths | # | Factor | Evidence | Impact | |---|--------|----------|--------| | S1 | {specific strength} | {supporting data} | High/Med/Low | | S2 | ... | ... | ... | ### Weaknesses | # | Factor | Evidence | Impact | |---|--------|----------|--------| | W1 | {specific weakness} | {supporting data} | High/Med/Low | | W2 | ... | ... | ... | ## External Factors ### Opportunities | # | Factor | Evidence | Likelihood | |---|--------|----------|-----------| | O1 | {specific opportunity} | {supporting data} | High/Med/Low | | O2 | ... | ... | ... | ### Threats | # | Factor | Evidence | Likelihood | |---|--------|----------|-----------| | T1 | {specific threat} | {supporting data} | High/Med/Low | | T2 | ... | ... | ... | ## TOWS Strategy Matrix | | Opportunities (O1, O2...) | Threats (T1, T2...) | |---|---|---| | **Strengths** | SO: {strategy using S to capture O} | ST: {strategy using S to counter T} | | **Weaknesses** | WO: {strategy to fix W and capture O} | WT: {strategy to minimize W and avoid T} | ## Recommended Actions 1. **{Strategy name}** — {description} (from: {SO/ST/WO/WT}) - Timeline: ... - Success metric: ... 2. ...
Scenario: SWOT for a mid-size Taiwanese bubble tea chain considering expansion to Japan
Analysis:
| Quadrant | Factor | Classification | Reasoning | |----------|--------|---------------|-----------| | S | Proprietary tapioca pearl recipe with unique texture | Internal ✓ | Company's own R&D product | | S | 50 stores in Taiwan with proven SOP | Internal ✓ | Company's operational asset | | W | No Japanese-speaking management team | Internal ✓ | Company's own resource gap | | W | Limited capital for overseas expansion | Internal ✓ | Company's financial constraint | | O | Japanese bubble tea market growing 15% YoY | External ✓ | Market trend, not company-controlled | | O | Japan-Taiwan tourism recovery post-COVID | External ✓ | Macro environment factor | | T | Established Japanese competitors (Gong Cha, CoCo) | External ✓ | Competitor landscape | | T | Yen depreciation increasing import costs | External ✓ | Economic condition |
TOWS strategy example:
Scenario: Same bubble tea chain
What went wrong:
| Factor | Placed In | Should Be | Why | |--------|-----------|-----------|-----| | "Japanese bubble tea market growing 15% YoY" | Strength ❌ | Opportunity | Market growth is an external trend — the company cannot control it. Violates Iron Law: internal factors only in S/W. | | "Our team lacks experience" | Threat ❌ | Weakness | Team composition is internal — the company controls who it hires. Violates Iron Law: external factors only in O/T. | | "Good brand" | Strength ⚠️ | Needs specificity | Too vague to be actionable. What makes the brand good? Recognition in Taiwan? Social media following? Specific awards? |
references/tows-case-studies.mdreferences/framework-comparison.md| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 33,097 | 32,123 | -3% | 1 | 1 | 0% | 5,796 | 7,035 | +21% | 0 | 0 | — |
case-02 | pass→pass | 19,395 | 16,703 | -14% | 1 | 1 | 0% | 3,144 | 4,829 | +54% | 0 | 0 | — |
case-03 | pass→pass | 21,182 | 23,309 | +10% | 1 | 1 | 0% | 3,295 | 5,334 | +62% | 0 | 0 | — |
case-04 | pass→fail | 13,805 | 13,155 | -5% | 1 | 1 | 0% | 2,334 | 4,420 | +89% | 0 | 0 | — |
case-05 | pass→pass | 14,981 | 17,671 | +18% | 1 | 1 | 0% | 2,268 | 5,334 | +135% | 0 | 0 | — |
case-06 | pass→pass | 14,139 | 17,937 | +27% | 1 | 1 | 0% | 2,320 | 5,180 | +123% | 0 | 0 | — |
case-07 | fail→pass | 9,790 | 14,469 | +48% | 1 | 1 | 0% | 1,461 | 4,559 | +212% | 0 | 0 | — |
case-08 | pass→pass | 9,491 | 8,278 | -13% | 1 | 1 | 0% | 1,399 | 3,342 | +139% | 0 | 0 | — |
case-09 | pass→pass | 14,830 | 16,702 | +13% | 1 | 1 | 0% | 2,428 | 4,679 | +93% | 0 | 0 | — |
case-10 | pass→fail | 17,380 | 24,172 | +39% | 1 | 1 | 0% | 2,568 | 5,702 | +122% | 0 | 0 | — |
case-11 | fail→pass | 14,709 | 17,273 | +17% | 1 | 1 | 0% | 2,414 | 5,247 | +117% | 0 | 0 | — |
case-12 | pass→pass | 11,857 | 10,800 | -9% | 1 | 1 | 0% | 1,963 | 3,969 | +102% | 0 | 0 | — |
case-18 | pass→pass | 12,886 | 22,226 | +72% | 1 | 1 | 0% | 2,149 | 5,378 | +150% | 0 | 0 | — |
case-13 | pass→pass | 8,139 | 6,208 | -24% | 1 | 1 | 0% | 1,390 | 3,224 | +132% | 0 | 0 | — |
case-14 | pass→pass | 10,593 | 12,090 | +14% | 1 | 1 | 0% | 1,700 | 3,844 | +126% | 0 | 0 | — |
case-15 | fail→pass | 26,002 | 13,859 | -47% | 1 | 1 | 0% | 4,194 | 4,460 | +6% | 0 | 0 | — |
case-16 | pass→pass | 16,122 | 22,030 | +37% | 1 | 1 | 0% | 2,587 | 5,914 | +129% | 0 | 0 | — |
case-17 | pass→pass | 22,540 | 26,511 | +18% | 1 | 1 | 0% | 3,718 | 6,709 | +80% | 0 | 0 | — |
case-19 | fail→pass | 16,736 | 17,390 | +4% | 1 | 1 | 0% | 2,277 | 4,720 | +107% | 0 | 0 | — |
case-20 | pass→pass | 22,744 | 24,817 | +9% | 1 | 1 | 0% | 3,345 | 6,554 | +96% | 0 | 0 | — |
case-21 | pass→pass | 16,094 | 22,722 | +41% | 1 | 1 | 0% | 2,344 | 5,483 | +134% | 0 | 0 | — |
case-22 | pass→pass | 10,876 | 16,389 | +51% | 1 | 1 | 0% | 1,845 | 4,939 | +168% | 0 | 0 | — |
case-23 | pass→pass | 12,670 | 15,122 | +19% | 1 | 1 | 0% | 2,037 | 4,403 | +116% | 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 +13 percentage points is the difference between those two pass rates over the 23 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.