| name | strategic-analysis |
| description | Apply structured strategic thinking before implementation. Use when facing architectural decisions, system design, refactoring, build-vs-buy choices, debugging complex issues, planning migrations, writing technical proposals, or any task where jumping straight to code would be premature. Triggers on: "design", "architect", "plan", "strategy", "evaluate", "decide", "trade-off", "should we", "how should", "what approach", "refactor", team dynamics questions, or when the problem has multiple valid solutions. Also activates on /strategize.
|
Strategic Analysis Framework
Apply all 5 lenses sequentially before proposing any solution. Each lens is a genuine reasoning step, not a checkbox.
Lens 1: Pattern Recognition
Identify what is actually happening before deciding what to do.
- What is the underlying trend in this codebase / system / problem?
- Why does this problem keep recurring, or why has it emerged now?
- What happens next if the current trajectory continues unchanged?
- Are there patterns from other projects, domains, or historical precedents that map to this situation?
Output a brief Patterns Observed section.
Lens 2: Root-Cause Questioning
Challenge the framing of the problem itself.
- What evidence do we have vs. what are we assuming?
- What is the real problem — not the symptom?
- What would need to change if we solved the root cause instead of patching?
- If starting from zero with no legacy constraints, how would we design this?
Output a brief Reframed Problem section that may differ from the original ask.
Lens 3: Resource & Impact Analysis
Understand the economics before committing resources.
- Where does the real value come from in this system / feature / decision?
- What drives cost and complexity (identify the variables, not the numbers)?
- How sustainable is the current approach at 2×, 10×, 100× scale?
- What metrics actually matter for this decision?
Output a brief Resource Reality section.
Lens 4: Hands-On Validation
Bias toward experiments over speculation.
- What is the smallest experiment that validates or kills the proposed approach?
- Define success criteria before starting.
- What is the fastest path to real-world feedback?
Output a brief Validation Plan with a concrete next step.
Lens 5: Clear Communication
Ensure the strategy can be understood and acted upon.
- What is the ONE thing the reader must take away?
- Can this be shown visually? (diagram, comparison table, decision tree)
- Who is the audience and what depth do they need?
Output a brief Key Message section and offer to generate a visual artifact.
Synthesis
- Strategic Recommendation — clear, concise position
- Confidence Level — High / Medium / Low with reasoning
- Key Risks — what could go wrong and how to mitigate
- Immediate Next Action — the single most important thing to do right now
Adaptive Depth
- Quick decisions (naming, small refactors, tool choice): 1–2 sentences per lens
- Medium decisions (API design, library choice, feature architecture): Full treatment, paragraph each
- High-stakes decisions (migrations, redesigns, platform bets): Deep analysis, include diagrams
Extended Frameworks
Available for deeper analysis when core lenses reveal conflicting signals:
- De Bono Six Thinking Hats →
references/six-hats.md — Team-oriented decisions, stakeholder conflict
- Islamic Epistemology Cognitive Levels →
references/cognitive-levels.md — Depth-of-understanding analysis
- Bridges Synthesis →
references/bridges-synthesis.md — Classical wisdom × modern engineering
- Ibn Khaldun's Muqaddimah →
references/ibn-khaldun.md — Team dynamics, organizational cohesion, decay signals
- Security Engineering (Anderson) →
references/security-engineering.md — 4-element security framework, threat modeling, usability, economics of security. Load when the decision or Foresight Report has security properties.
Invoke extended frameworks when:
- Core lenses reveal conflicting signals
- The problem has team/organizational dimensions
- The problem has philosophical or ethical dimensions
- The user invokes
/istishraf:think-deeper or /istishraf:think-deeper --all