SWOT Analysis · Strategy field boardFree · MIT · Base Skill #298

See where you stand before you choose where to go.

Give your assistant the facts about a product or business. It returns an honest four-quadrant SWOT, cross-references the quadrants, and turns them into 3–5 prioritized moves with owners and metrics.

Original instructions by Pawel Huryn (phuryn/pm-skills), preserved byte for byte.

HelpfulHarmful Internal
Strengths
  • Clear fit with 20–45 person studios
  • $11.50 vs $16 hot catering (28% gap)
Weaknesses
  • 6% monthly churn ≈ 52% of offices a year
  • 31% margin vs 38% target
External
Opportunities
  • Underserved mid-sized offices
  • Flex headcount for hybrid teams
Threats
  • Occupancy at ~55% of pre-2020
  • Packaging rule and wage costs
Abridged from a saved run on “Tidewell Desk”, an invented lunch-kit startup. The full output has seven items per quadrant.
5–7items per quadrant
4cross-reference pairs
3–5strategic moves
0API keys needed

Counts are rules written in the Skill’s instructions, not performance claims.


What it is

A single instruction file that turns a capable AI assistant into a careful strategic analyst. It separates internal factors (strengths, weaknesses) from external ones (opportunities, threats), asks for honesty about gaps, and requires threats to be rated for probability and impact.

The analysis does not stop at four lists. The Skill makes the assistant cross-reference them, propose recommendations, assign owners, name metrics and choose a stance: Build, Defend, Pivot or Exit.

Who it is for

  • Product managerspreparing a roadmap review or a launch decision.
  • Founders and operatorspressure-testing an expansion or a new market.
  • Strategy and consulting teamsneeding a structured first draft fast.
  • Students and analystslearning to apply SWOT rigorously.

Works well for

  • Quarterly strategy reviews and roadmap planning
  • Go / no-go decisions such as a new city or segment
  • Competitive positioning against named rivals
  • Partnership, pricing or resource-allocation discussions

Not designed for

  • Financial forecasts, valuations or investment advice
  • Live market data — it uses what you supply or what your assistant can look up
  • Replacing customer research or due diligence
  • Legal, regulatory or compliance determinations

What you provide, what you receive

You provide

  • Product description and current state
  • Competitive landscape and market context
  • Capabilities, resources and constraints
  • Market trends and industry dynamics
  • Customer feedback or usage data (optional)

You receive

  • 5–7 items in each quadrant, threats rated for probability and impact
  • Cross-reference: S×O, W×O, S×T, W×T
  • 3–5 strategic recommendations
  • Prioritized actions with owners and timing
  • Metrics to track and a Build / Defend / Pivot / Exit stance

Where it runs

Any AI assistant that can read a Skill file, including Claude Code and Codex. Download the free plaintext package from Finch, then install it with your assistant’s Skill method or ask the assistant to read SKILL.md.

No extra API key, script or service. You still need access to an assistant model, which may have its own subscription or usage cost. The Skill does not browse the web on its own; if you want fresh market facts, supply them or let an assistant with search tools gather them.

Limits and licence

The output is a structured judgment, only as good as the facts supplied. Review every claim before using it in a decision.

Licensed under MIT. Copyright (c) 2026 Pawel Huryn. Source: phuryn/pm-skills, pinned at commit 2a12295a7717, path pm-product-strategy/skills/swot-analysis. The package adds a README, NOTICE, provenance note and a promotional cover; the instructions are unchanged.


How it was checked

One controlled run on a fictional company produced all nine required parts of the analysis, labelled every assumption and derived figure, and listed what was unknown. The published package was downloaded back from Finch and matched the audited files byte for byte. Native installation inside each assistant client and long-term business outcomes were not tested. See the Benchmark tab for the full evidence trail.