$npx -y skills add stevembarclay/pencilplaybook --skill best-in-world-strategyExcellence-first strategic decision support. This skill should be used when users need to choose between options, pressure-test a plan, evaluate risk, or make a specific decision — across security, product, growth, operations, org design, and finance. Use when there is a decision
| 1 | # Best-In-World Strategy |
| 2 | |
| 3 | Set an excellence baseline first. Adapt intentionally second. |
| 4 | |
| 5 | ## Modes |
| 6 | |
| 7 | Use `Speed` mode by default. Reserve `Full` mode for explicitly high-stakes or complex decisions. |
| 8 | |
| 9 | - `Speed` (default): frame the decision → world-class baseline → recommendation → downside controls. Rigorous but concise. No options scoring, no red-team unless the stakes clearly warrant it. |
| 10 | - `Full` (when user asks for comprehensive analysis, or the decision is clearly high-stakes / irreversible): complete 9-step workflow with constraint ledger, options scoring, red-team, and learning loop. |
| 11 | - `Evidence-Backed` modifier (optional): add external citations to either mode when user explicitly asks. |
| 12 | |
| 13 | Switch to `Risk-Acceptance Override` when user states willingness to accept risk and move quickly — drops options scoring, keeps assumptions + failure modes + downside controls. |
| 14 | |
| 15 | ## Workflow |
| 16 | |
| 17 | 1. Frame the decision. |
| 18 | - Write one decision sentence. |
| 19 | - Capture decision type: `architecture`, `security/risk`, `product`, `go-to-market`, `operations`, `org/talent`, `capital/finance`. |
| 20 | - Capture objective, owner, time horizon, and success metric. |
| 21 | |
| 22 | 2. Build a constraint ledger. |
| 23 | - Separate `hard constraints` from `soft constraints`. |
| 24 | - Record budget, timeline, legal, staffing, and political constraints. |
| 25 | |
| 26 | 3. Set the best-in-world baseline. |
| 27 | - Identify 2-4 expert archetypes relevant to the decision. |
| 28 | - For each archetype, output: |
| 29 | - `principle` |
| 30 | - `expected benefit` |
| 31 | - `cost/complexity` |
| 32 | - `confidence` |
| 33 | - Mark each item as `non-negotiable` or `adaptable`. |
| 34 | |
| 35 | 4. Apply first principles. |
| 36 | - Separate `facts`, `assumptions`, and `conventions`. |
| 37 | - State must-be-true conditions. |
| 38 | - Isolate the smallest causal drivers. |
| 39 | |
| 40 | 5. Run red-team pressure testing. |
| 41 | - Ask hard skeptical questions. |
| 42 | - Surface first-order and second-order failure modes. |
| 43 | - Define one disconfirming test. |
| 44 | |
| 45 | 6. Generate options. |
| 46 | - Option A: world-class ideal. |
| 47 | - Option B: pragmatic high-confidence. |
| 48 | - Option C: minimum viable risk-controlled. |
| 49 | |
| 50 | 7. Score options. |
| 51 | - Score each option 1-5 on: |
| 52 | - `impact` |
| 53 | - `risk` |
| 54 | - `cost` |
| 55 | - `speed` |
| 56 | - `reversibility` |
| 57 | - `learning value` |
| 58 | - Use default weights unless user provides custom weights: |
| 59 | - `impact 30%` |
| 60 | - `risk 20%` |
| 61 | - `cost 15%` |
| 62 | - `speed 15%` |
| 63 | - `reversibility 10%` |
| 64 | - `learning value 10%` |
| 65 | |
| 66 | 8. Recommend and sequence. |
| 67 | - Recommend one option and explain the tradeoffs. |
| 68 | - Sequence as `now`, `next`, `later`. |
| 69 | - Mark major decisions as `two-way door` or `one-way door`. |
| 70 | - Define downside controls: |
| 71 | - max acceptable loss |
| 72 | - kill criteria |
| 73 | - revisit trigger/date |
| 74 | - Include 2-3 no-regret moves. |
| 75 | |
| 76 | 9. Define learning loop. |
| 77 | - Define what to measure. |
| 78 | - Define what would change the current strategy. |
| 79 | |
| 80 | ## Risk-Acceptance Override |
| 81 | |
| 82 | When user accepts risk explicitly: |
| 83 | |
| 84 | 1. Keep rigor floor. |
| 85 | 2. Switch to speed output. |
| 86 | 3. Include this line exactly: |
| 87 | `Decision owner accepts risk and prefers speed over additional validation at this stage.` |
| 88 | 4. Keep mandatory safeguards: |
| 89 | - assumptions list |
| 90 | - confidence tags (`high`, `medium`, `low`) |
| 91 | - top failure modes |
| 92 | - one disconfirming test |
| 93 | - max acceptable loss |
| 94 | - kill criteria |
| 95 | - revisit trigger/date |
| 96 | |
| 97 | ## Output Contract (Speed — default) |
| 98 | |
| 99 | ```markdown |
| 100 | ## Decision |
| 101 | [One sentence] |
| 102 | |
| 103 | ## What the Best in the World Would Do |
| 104 | [2–3 bullets. Named practitioners or archetypes. Specific practices, not principles.] |
| 105 | |
| 106 | ## Recommendation |
| 107 | [One option. Why this one. What's intentionally deferred.] |
| 108 | |
| 109 | ## Key Assumptions |
| 110 | [2–3 items with confidence tags: high / medium / low] |
| 111 | |
| 112 | ## Downside Controls |
| 113 | - Max acceptable loss: |
| 114 | - Kill criteria: |
| 115 | - Revisit trigger/date: |
| 116 | ``` |
| 117 | |
| 118 | ## Output Contract (Full) |
| 119 | |
| 120 | ```markdown |
| 121 | ## Decision |
| 122 | |
| 123 | ## Mode |
| 124 | - Selected mode: |
| 125 | - Risk-Acceptance Override: |
| 126 | |
| 127 | ## Decision Type |
| 128 | |
| 129 | ## Objective and Success Metric |
| 130 | - Objective: |
| 131 | - Success metric: |
| 132 | - Time horizon: |
| 133 | - Owner: |
| 134 | |
| 135 | ## Constraint Ledger |
| 136 | - Hard constraints: |
| 137 | - Soft constraints: |
| 138 | |
| 139 | ## What the Best in the World Would Say |
| 140 | - Expert archetype 1: |
| 141 | - Principle: |
| 142 | - Expected benefit: |
| 143 | - Cost/complexity: |
| 144 | - Confidence: |
| 145 | - Expert archetype 2: |
| 146 | - Principle: |
| 147 | - Expected benefit: |
| 148 | - Cost/complexity: |
| 149 | - Confidence: |
| 150 | - Non-negotiables: |
| 151 | - Adaptable elements: |
| 152 | |
| 153 | ## First-Principles Breakdown |
| 154 | - Facts: |
| 155 | - Assumptions: |
| 156 | - Conventions: |
| 157 | - Must-be-true conditions: |
| 158 | - Causal drivers: |
| 159 | |
| 160 | ## Red-Team Questions and Failure Modes |
| 161 | - Hard questions: |
| 162 | - First-order failures: |
| 163 | - Second-order failures: |
| 164 | - Disconfirming test: |
| 165 | |
| 166 | ## Strategic Options |
| 167 | - Option A (world-class ideal): |
| 168 | - Option B (pragmatic high-confidenc |