natan-mohart/24-strategy-skills-for-claude

24 Agent Skills for Claude, one folder per skill, each a SKILL.md plus, for the quantitative ones, a bundled Python calculator. Built around McKinsey’s actual documented problem-solving method (define the problem, build a MECE issue tree, form hypotheses, prioritize, build a workplan, analyze, synthesize, communicate) and their real practice-area names, instead of a generic “diagnose → communicate” template.

13 of the 24 skills ship a tested Python calculator - NPV/IRR, LTV:CAC, TAM/SAM/SOM, Van Westendorp pricing, spans and layers, GE-McKinsey 9-box, decision-tree expected value, risk scoring, BATNA/ZOPA, RICE, and synergy NPV, so the number-heavy skills compute a real answer instead of describing a method and leaving the math to you.

Contributions welcome! Found a gap, a calculator bug, or have a skill to add? See Contributing.

What are Skills?

Skills are markdown files that give Claude specialized knowledge and a repeatable method for a specific task. When Claude has access to these, it recognizes when a conversation matches a skill’s description and applies that skill’s method automatically, no need to invoke it by name, though you can (Use the market-mapping skill to size this opportunity).

How the skills fit together

situation-assessment and assumption-audit sit at the start of any real engagement. full-potential-diagnostic and business-case-builder size what a decision is worth in dollars. value-realization is the only skill in the pack that looks backward, it closes the loop by checking what the projection-producing skills predicted against what actually happened.

                    ┌───────────────────────────────────────┐
                    │         situation-assessment           │
                    │   (hypothesis tree + fact base first)  │
                    └───────────────────┬─────────────────────┘
                                        │
   ┌─────────────┬──────────────┬───────┼────────┬──────────────┬──────────────┐
   ▼             ▼              ▼       ▼        ▼              ▼              ▼
┌─────────┐ ┌──────────┐ ┌────────────┐ ┌──────────────┐ ┌───────────┐ ┌──────────────┐
│Problem  │ │Market &  │ │Strategy &  │ │Performance   │ │Risk &     │ │Communication │
│Def.     │ │Competitive│ │Corp Finance│ │Transformation│ │Value Cap. │ │& Change      │
├─────────┤ ├──────────┤ ├────────────┤ ├──────────────┤ ├───────────┤ ├──────────────┤
│growth-  │ │market-   │ │strategic-  │ │operating-    │ │war-gaming │ │stakeholder-  │
│ barriers│ │ mapping  │ │ options    │ │ model-design │ │risk-      │ │ alignment    │
│full-    │ │competitiv│ │pricing-    │ │spans-layers  │ │ mitigation│ │narrative-    │
│ potential│ │e-intel   │ │ strategy   │ │initiative-   │ │synergy-   │ │ builder      │
│         │ │customer- │ │business-   │ │ prioritizer  │ │ case      │ │decision-memo │
│         │ │ segment  │ │ case       │ │100-day-plan  │ │value-     │ │negotiation-  │
│         │ │unit-econ │ │capital-    │ │              │ │ realization│ │ prep         │
│         │ │          │ │ realloc    │ │              │ │(checks →) │ │              │
└─────────┘ └──────────┘ └─────┬──────┘ └──────┬───────┘ └─────┬─────┘ └──────────────┘
                               │               │                │
                               └───────────────┴────────────────┘
                     value-realization checks these three against actuals

Available Skills

🧮 = ships a bundled Python calculator in scripts/

Skill Categories

01 · Problem Definition

Before anyone touches a solution. situation-assessment · growth-barriers · assumption-audit · full-potential-diagnostic

02 · Market & Competitive Analysis

Where the value actually sits. market-mapping · competitive-intel · customer-segmentation · unit-economics-health-check

03 · Strategy & Corporate Finance

Where bets and capital get decided. McKinsey’s own practice name. strategic-options · pricing-strategy · business-case-builder · capital-reallocation-review

04 · Performance Transformation

Turning strategy into delivered results. The other real McKinsey practice name. operating-model-design · spans-layers-org-design · initiative-prioritizer · 100-day-plan

05 · Risk & Value Capture

Before it launches, and after. war-gaming · risk-mitigation · synergy-case-builder · value-realization

06 · Communication & Change

Making the work survive the meeting. stakeholder-alignment · narrative-builder · decision-memo · negotiation-prep

Calculators

Every calculator was run against a worked test case while this pack was built, the numbers in each skill’s Example section are real output, not invented illustrations. A few examples of what they catch that a prose-only skill can’t:

• market-mapping/scripts/tam_sam_som.py flags when a top-down and bottom-up market-size estimate disagree by more than 30%
• business-case-builder/scripts/npv_case.py returns a full revenue × discount-rate NPV sensitivity grid, not a single point estimate
• war-gaming/scripts/decision_tree_ev.py flags when the highest-expected-value option and the safest-worst-case option are different choices
• stakeholder-alignment/scripts/power_interest_grid.py computes coalition math by power, not headcount, and flags when opposition outweighs support
• negotiation-prep/scripts/batna_zopa.py tells you plainly when no Zone of Possible Agreement exists, instead of letting you negotiate on a gap that can’t close

Installation

Option 1: Claude.ai / Claude Desktop

Settings → Skills → Add → Upload a skill → select a skill’s SKILL.md (or its folder, where supported). Repeat per skill, or upload all 24.

Option 2: Claude Code — CLI install

npx skills add Natan-Mohart/24-strategy-skills-for-claude

Installs into .claude/skills/ (project) or ~/.claude/skills/ (personal) depending on how you answer the CLI prompt. Claude Code picks up SKILL.md files automatically in the next session.

Option 3: Claude Code — clone and copy

git clone https://github.com/Natan-Mohart/24-strategy-skills-for-claude.git
cp -r 24-strategy-skills-for-claude/skills/* ~/.claude/skills/

Option 4: Fork and customize

Fork this repo, edit any SKILL.md or scripts/*.py to fit your own business, then install from your fork using either option above.

Skill format

Every SKILL.md follows the same structure: When to use → What it does → Method → Inputs → Output format → Example → Common pitfalls, grounded in named frameworks (MECE, RAPID/DACI, Theory of Constraints, Van Westendorp, GE-McKinsey, RICE, Jobs-to-be-Done, real options, BATNA/ZOPA, spans and layers) instead of generic “brainstorm some ideas” instructions.

Why this exists

Most “strategy skills for Claude” packs mirror the same six-phase consulting narrative with prose-only prompting frameworks and an identical topic list. This one:

• Uses McKinsey’s actual documented hypothesis-driven method and real practice-area names
• Runs situation assessment and hypothesis-tree building as one skill instead of two overlapping ones, hypothesis on day one, facts that test it second
• Includes topics a typical strategy-skill pack skips entirely: full-potential diagnostics, 100-day plans, M&A synergy cases, spans-and-layers org design, unit economics health checks, and negotiation prep
• Ships 13 tested Python calculators instead of zero
• Closes the loop: value-realization checks the projections that business-case-builder, full-potential-diagnostic, and synergy-case-builder produce against what actually happened

Contributing

Found a gap, a bug in a calculator, or a skill that overlaps with another? See CONTRIBUTING.md.

License

MIT - use it, fork it, adapt it.

---