Strategic Thinking

strategy is choosing what you don't do as much as what you do. end with explicit bets and guardrails.

intent

this skill frames decisions under constraint by connecting intent, context, capabilities, and options into coherent bets, tradeoffs, risks, and a sequenced path. use it when the user asks for strategic thinking (including obvious misspellings of "strategy"), needs direction on competitive positioning, roadmap narrative, where to play and how to win, portfolio prioritization, or leadership narrative that links goals to constraints. skip this skill if the request is purely step-by-step implementation detail with no strategic framing or choice among directions.

inputs

user context required:

• strategic question (one sentence, e.g. "how should we win in X given Y?")
• goals and non-negotiables (if missing, skill will ask up to 3 clarifying questions)
• constraints (budget, timeline, team, market access, regulatory)
• current capabilities and assets (team skills, tech stack, brand, distribution, data)
• competitive or market context (if thin, skill will note the gap plainly)

optional trigger inputs:

• request for memo or deck storyline (adds "short story" subsection after risks & cadence)
• existing options or bets to evaluate (skill will map against phases)

external connections:

• none required. skill works with provided context and conversation history.

edge cases:

• if strategic question is vague or missing, skill pauses at setup and asks clarifying questions
• if goals and constraints conflict, skill names the conflict explicitly and may propose phased sequencing
• if no credible advantage exists, skill pivots to "options to build advantage"
• if user has insufficient information for landscape analysis, skill states what is unknown instead of fabricating market facts

procedure

step 1: setup (run before starting)

in one short block:

1. state the strategic question in one sentence (e.g. "how should we win in X given Y?")
2. confirm you will proceed through intent > landscape > advantage > options > choice > risks & cadence (state this line)
3. if goals, constraints, or non-negotiables are missing, ask at most 3 questions in one message, then proceed
4. note any remaining gaps or working guesses in plain language (no bracket tags in setup)
5. if the user asked for a memo or deck storyline, note that "short story" subsection will be added after risks & cadence

input: user's strategic request and context output: confirmed strategic question, list of gaps (if any), setup summary

step 2: intent

frame the win definition and non-goals.

1. state what outcome you are optimizing for and in what timeframe (the win definition)
2. state explicitly what is out of scope or explicitly rejected (non-goals)
3. name any leadership narrative that connects the win to broader business context

input: setup summary, user's stated goals output: 2-4 sentences clarifying win definition and non-goals

step 3: landscape

map forces that matter and their implications.

1. list forces by category: customers, competition, technology, regulation, economics, or internal (team, org structure)
2. for each force, state the implication (what it means for your move) as a short bullet, not an encyclopedic description
3. if information is thin, say plainly what is unknown

input: user context on market, competitors, customer needs, internal state output: 3-7 implication bullets per category

step 4: advantage (or honest gap)

identify where durable advantage could come from.

1. scan these sources: assets (team, tech, brand), learning loops (data feedback, iteration speed), distribution (channels, partnerships), cost structure, network effects
2. for each potential source, state whether it is credible and defensible over 12-24 months
3. if no credible advantage is visible, say so and pivot to "options to build advantage"

input: landscape, user's capabilities and assets output: 3-5 sentences naming one or two advantage sources, or explicit statement that none is credible yet

step 5: options

generate 2-4 mutually distinct strategies or postures.

1. name each option in one short phrase (e.g. "land grab", "build moat", "sell at scale")
2. for each option, fill in all four fields:
   • bet: what you are betting will happen (customer behavior, market evolution, tech feasibility)
   • cost: rough resource commitment (headcount, cash, time-to-market)
   • kill signal: the metric or event that forces you to abandon this option
3. ensure each option is genuinely distinct; avoid near-duplicates

input: landscape, advantage analysis, strategic question output: 2-4 option blocks, each with bet, cost, kill signal

step 6: choice

name the primary option and explain why now and what you defer or reject.

1. state which option (by name from step 5) is the primary bet
2. if multiple parallel bets are justified, name them and explain why they are not sequential
3. state explicitly what you are deferring or rejecting and why (e.g. "we are not pursuing option X because Y")
4. explain why now (what has changed, or what is the forcing event)

input: options block, landscape, intent output: 2-4 sentences naming choice, explaining why now, stating deferrals

step 7: risks & cadence

surface top risks with mitigations and set review cadence.

1. list top 3 risks (not every risk, the ones that could kill the bet)
2. for each risk, state a mitigation or test (not a platitude; something actionable)
3. state the 90-day focus (what has to be true in 90 days)
4. state the 12-month thesis (what has to be true for the bet to be "right")
5. state the review trigger (metric or event that forces strategic rethink)

input: choice, landscape, execution constraints output: risk + mitigation bullets, 90-day focus, 12-month thesis, review trigger

step 8 (optional): short story

if user requested a memo or deck storyline, add this subsection after risks & cadence.

1. write 5-7 sentences that flow: tension (what is the problem or choice) > insight (what you learned in the analysis) > decision (the choice and why) > proof path (how you will know if it is right)
2. use narrative voice, not bullets; this is for leadership or board communication

input: all prior phases output: 5-7 sentence narrative block

decision points

• if goals, constraints, or non-negotiables are missing: ask up to 3 clarifying questions in one message, then proceed with stated working assumptions. do not stall the analysis.

• if strategic question is vague or abstract: propose a concrete one-sentence reframing and confirm before proceeding.

• if goals and constraints conflict: name the conflict explicitly (e.g. "you want 90-day launch but also zero technical debt", which is not credible). propose phased sequencing or ask which constraint wins.

• if no credible advantage exists: say so plainly. pivot options to focus on "ways to build advantage" (e.g. acquire data, hire talent, partner for distribution) instead of ways to exploit existing advantage.

• if information is thin (market size, customer behavior, competitive moves): state plainly what is unknown. do not fabricate market facts. offer to refine the analysis if user provides more data.

• if user requested a memo or deck: add the "short story" subsection after risks & cadence. otherwise, omit it.

• if choice introduces a new strategy not in the options block: label it explicitly as a revision and explain why. do not let it sneak in without naming it.

output contract

skill output is a formatted text response with these sections in order, each clearly headed:

1. setup (one short block confirming strategic question, default pass, any gaps, optional short story flag)
2. intent (win definition and non-goals, 2-4 sentences)
3. landscape (forces and implications, 3-7 bullets per category)
4. advantage (or "honest gap", 3-5 sentences identifying defensible advantage sources or pivoting to build)
5. options (2-4 mutually distinct options, each with bet, cost, kill signal)
6. choice (1-2 sentences naming primary option, why now, deferrals)
7. risks & cadence (3 top risks with mitigations, 90-day focus, 12-month thesis, review trigger)
8. short story (optional, 5-7 sentences if user requested memo or deck narrative)

no frontmatter, no code blocks, no meta-commentary. plain markdown headings and bullets.

outcome signal

the user knows the skill worked when they can:

• state the strategic question and see it reflected back with clarity
• name the win definition and non-goals without ambiguity
• point to landscape forces that actually matter in their context
• identify which option they are betting on and why competing options were rejected
• state the top 3 risks and what would trigger a rethink
• use the 90-day focus as a checkpoint for roadmap planning
• (if requested) send the "short story" to leadership or board without rewriting it

skill fails if the analysis is generic (e.g. "be customer-centric and innovate"), if choice is not grounded in options, or if risks lack actionable mitigations.

---

credits: original skill authored by ysskrishna (clawhub). enriched per implexa standards.