Strategy & decision-making

• decision-matrix — Compare options against weighted criteria to pick defensibly.
• expected-value — Weight outcomes by probability to choose under uncertainty.
• bayesian-reasoning-calibration — Update probabilities with evidence and calibrate confidence.
• heuristics-and-checklists — Prevent recurring errors with fast rules and 5–9 item checklists.
• hypotheticals-counterfactuals — Stress-test decisions with "what if" scenarios and premortems.
• kill-criteria-exit-ramps — Define when to stop a project before sunk cost traps you.
• alignment-values-north-star — Connect daily choices to a shared North Star and values.
• environmental-scanning-foresight — Spot weak signals and plan scenarios via PESTLE.
• portfolio-roadmapping-bets — Size and sequence H1/H2/H3 bets with 70-20-10 balance.
• prioritization-effort-impact — Sort a backlog into Quick Wins, Big Bets, Time Sinks.
• project-risk-register — Identify, score, and monitor project risks by probability × impact.
• prototyping-pretotyping — Validate demand cheaply before building via fake doors and MVPs.
• postmortem — Turn failures into learning with blameless timelines and SMART actions.
• strategy-and-competitive-analysis — Build strategy using Rumelt's kernel, Porter, Blue Ocean, Playing to Win.

Problem-solving & analysis

• decomposition-reconstruction — Break complex systems into parts, then recompose with insight.
• causal-inference-root-cause — Distinguish causation from correlation; find true root causes.
• abstraction-concrete-examples — Move between abstract principle and concrete example to clarify.
• layered-reasoning — Reason consistently across strategic / tactical / operational layers.
• negative-contrastive-framing — Define fuzzy criteria by showing what they are NOT.
• synthesis-and-analogy — Integrate sources and transfer insights across domains.
• systems-thinking-leverage — Find leverage points in feedback loops and system archetypes.

Estimation & forecasting

• estimation-fermi — Produce order-of-magnitude estimates via decomposition and bounding.
• reference-class-forecasting — Anchor predictions in base rates from similar past events.
• forecast-premortem — Stress-test a forecast by assuming it failed; surface tail risks.
• scout-mindset-bias-check — Detect and remove bias with reversal tests and confidence audits.
• chain-estimation-decision-storytelling — Chain Fermi estimation → EV → narrative into a recommendation.

Dialogue & deliberation

• deliberation-debate-red-teaming — Stress-test plans via Oxford debate, fishbowl, devil's advocate.
• dialectical-mapping-steelmanning — Steelman both sides and synthesize a principled third way.
• chain-roleplay-debate-synthesis — Multi-expert roleplay and debate for complex decisions.
• role-switch — Analyze decisions from each stakeholder's perspective to find alignment.

Ideation & creativity

• brainstorm-diverge-converge — Generate many ideas, cluster, then narrow via dot voting.
• constraint-based-creativity — Turn limitations into creative fuel via strategic constraints.
• morphological-analysis-triz — Systematically explore design space and resolve contradictions.

Productivity & learning

• focus-timeboxing-8020 — Protect attention with Pareto, Pomodoro, and deep-work blocks.
• memory-retrieval-learning — Build durable retention via spaced repetition and retrieval practice.
• socratic-teaching-scaffolds — Teach via strategic questions and fading scaffolds.
• meta-prompt-engineering — Turn vague prompts into reliable, structured, verifiable ones.

• discovery-interviews-surveys — Run unbiased JTBD interviews and surveys with thematic coding.
• design-of-experiments — Design rigorous factorial, RSM, and Taguchi experiments.
• domain-research-health-science — Formulate PICOT questions and GRADE clinical evidence.
• research-claim-map — Verify claims via triangulation, source grading, confidence calibration.
• ethics-safety-impact — Assess harms, fairness, and mitigations across stakeholders.
• evaluation-rubrics — Design rubrics with calibrated scales and inter-rater reliability.

General writing

• writing-structure-planner — Plan writing architecture via McPhee's structural diagramming.
• writing-revision — Run the three-pass revision: clutter, cognitive load, rhythm.
• writing-stickiness — Make messages memorable using the Heath SUCCESs framework.
• writing-pre-publish-checklist — Final 6-section quality gate before publishing.
• communication-storytelling — Craft narratives using arcs, tension, and audience framing.
• translation-reframing-audience-shift — Adapt content for a new audience without losing accuracy.
• one-pager-prd — Write concise one-pagers and PRDs for stakeholder alignment.

Scientific & academic writing

• scientific-manuscript-review — Edit IMRaD research manuscripts for clarity and rigor.
• grant-proposal-assistant — Assist NIH / NSF grant writing with Specific Aims + Strategy.
• academic-letter-architect — Craft evidence-based recommendation and nomination letters.
• scientific-email-polishing — Polish editor letters, reviewer responses, collaboration emails.
• scientific-clarity-checker — Audit scientific claims, hedging, and terminology consistency.
• career-document-architect — Develop research, teaching, diversity statements and CVs.

Architecture & product

• adr-architecture — Document architecture decisions with context, options, consequences.
• chain-spec-risk-metrics — Chain spec → risk → success metrics for launches.
• roadmap-backcast — Plan backward from deadline to identify critical path and feasibility.
• metrics-tree — Decompose a North Star metric into sub-metrics and leading indicators.

Design & information architecture

• information-architecture — Organize content via card sort, taxonomy, and tree testing.
• cognitive-design — Ground design decisions in perception, memory, and Gestalt.
• design-evaluation-audit — Audit existing designs against cognitive-alignment checklists.
• cognitive-fallacies-guard — Detect chartjunk, misleading axes, and bias in visualizations.
• visual-storytelling-design — Turn data into narrative via annotation and scrollytelling.
• d3-visualization — Build custom interactive charts, networks, and maps with D3.js.

Security

• security-threat-model — Apply STRIDE with trust boundaries and defense-in-depth.

Org & facilitation

• facilitation-patterns — Run productive meetings, workshops, and retros.
• negotiation-alignment-governance — Align stakeholders via RACI, BATNA, and conflict protocols.
• reviews-retros-reflection — Run blameless retros with SMART actions (Start/Stop/Continue, Sailboat).
• stakeholders-org-design — Map stakeholders; design teams via Conway and Team Topologies.

Data modeling & visualization

• data-schema-knowledge-modeling — Model entities and relationships for DBs, graphs, ontologies.
• code-data-analysis-scaffolds — Generate code scaffolds for analysis across Python / R / SQL.
• visualization-choice-reporting — Pick the right chart and narrate the insight to action.
• mapping-visualization-scaffolds — Map relationships with trees, networks, layers, swimlanes.

Geometric deep learning

• symmetry-discovery-questionnaire — Discover hidden data symmetries collaboratively.
• symmetry-group-identifier — Map symmetries to cyclic, dihedral, SO(3), SE(3), E(3) groups.
• symmetry-validation-suite — Empirically test invariance and equivariance claims.
• equivariant-architecture-designer — Design G-CNN, steerable, and e3nn architectures.
• model-equivariance-auditor — Verify implemented models respect their intended symmetries.

GraphRAG

• knowledge-graph-construction — Design LPG / RDF knowledge graphs from unstructured data.
• embedding-fusion-strategy — Combine semantic and structural embeddings for retrieval.
• retrieval-search-orchestration — Design graph + vector retrieval pipelines with provenance.
• graphrag-system-design — Architect end-to-end GraphRAG from pattern to deployment.
• graphrag-evaluation — Evaluate GraphRAG across KG quality, retrieval, hallucination.

Corporate finance & valuation (11)

Based on Damodaran's valuation curriculum.

• business-narrative-builder — Link a company's story to valuation drivers.
• financial-statement-analyzer — Normalize financials; compute FCFF, FCFE, key ratios.
• cost-of-capital-estimator — Compute WACC via CAPM, bottom-up beta, country risk.
• intrinsic-valuation-dcf — Run DDM / FCFE / FCFF DCF with terminal value and sensitivity.
• relative-valuation-multiples — Value via PE, PBV, EV/EBITDA vs peers or regression.
• capital-structure-optimizer — Find the debt ratio that minimizes WACC.
• project-investment-analyzer — Evaluate projects via NPV, IRR, EVA vs hurdle rate.
• dividend-buyback-analyzer — Decide dividends vs buybacks vs retained cash.
• special-situations-valuation — Value distressed, private, high-growth, and financial firms.
• valuation-reconciler — Reconcile DCF and multiples into a buy / sell / hold call.
• financial-unit-economics — Analyze CAC, LTV, contribution margin, cohort payback.

Household finance (12)

PDF-driven personal-finance pipeline. Pairs with the orchestration runtime at ~/Documents/Projects/financialplanning/ (inbox, archive, canonical JSON store, prompts, weekly static HTML dashboards). Powers the household-cfo agent and 8 specialists.

• pdf-statement-parser — Parse a single bank / brokerage / 401k / HSA / mortgage / tax PDF into normalized JSON with confidence.
• transaction-categorizer — Rules-then-LLM categorization with merchant normalization and rule learning.
• transaction-deduplicator — Composite-key dedupe across overlapping statement imports; preserves legitimate same-day repeats.
• recurring-charge-detector — Cluster same-merchant charges by cadence; promote at ≥3 occurrences; track dormant.
• statement-reconciler — Verify opening + Σ = closing; diagnose sign flips, missing rows, double-counts.
• cash-flow-forecaster — 60-day daily projection per cash account with trough confidence band.
• category-trend-analyzer — Current vs 6-month rolling vs YoY vs budget; outliers ranked by dollar impact.
• anomaly-fraud-scanner — 5-rule fraud / anomaly scan with severity, evidence, recommended action.
• portfolio-drift-rebalancer — Aggregate allocation across taxable + 401k + HSA; tax-efficient rebalance proposal.
• tax-loss-harvest-scanner — TLH candidates with wash-sale guard across all household accounts including spousal.
• hsa-receipt-vault — Track HSA-qualified out-of-pocket expenses for deferred tax-free reimbursement.
• household-finance-dashboard-builder — Single self-contained static HTML dashboard from the JSON store.

Game theory & strategic competition (8)

Domain-neutral primitives — portable across fantasy sports, poker, auctions, M&A.

• auction-first-price-shading — Optimal shaded bid for sealed first-price auctions.
• auction-winners-curse-haircut — Bayesian haircut for common-value auctions.
• adverse-selection-prior — Prior that an offered trade is +EV given they proposed it.
• variance-strategy-selector — Pick variance-seeking or minimizing posture by win probability.
• opponent-archetype-classifier — Bayesian archetype inference from observed behavior.
• matchup-win-probability-sim — P(win K of N cats) via Monte-Carlo or Poisson-binomial.
• category-allocation-best-response — Push / concede / contest allocation for H2H Categories.
• market-mechanics-betting — Turn beliefs into bets: Kelly sizing, Brier optimization.

Yahoo Fantasy Baseball (14)

Baseball-specific skills for a H2H Categories league. Pairs with the companion runtime at ~/Documents/Projects/yahoo-mlb/.

• mlb-league-state-reader — Parse Yahoo league state via Claude-in-Chrome.
• mlb-player-analyzer — Deep-dive a player across FanGraphs, Savant, RotoWire.
• mlb-matchup-analyzer — Grade a single game for fantasy impact (park, weather, SP).
• mlb-category-state-analyzer — Compute weekly cat state across all 10 categories.
• mlb-regression-flagger — Flag buy-low / sell-high via xwOBA / FIP vs surface stats.
• mlb-two-start-scout — Find two-start SPs and grade for QS scoring.
• mlb-closer-tracker — Track closer roles and handcuffs across all 30 teams.
• mlb-faab-sizer — Size FAAB bids; delegates to auction skills for shading.
• mlb-trade-evaluator — Score a trade across all 10 cats with adverse-selection prior.
• mlb-playoff-scheduler — Count games and grade matchups for weeks 21–23.
• mlb-opponent-profiler — Weekly archetype refresh for the 11 league opponents.
• mlb-signal-emitter — Validate and persist signals with schema enforcement.
• mlb-decision-logger — Append structured decision entries; run Monday calibration.
• mlb-beginner-translator — Wrap jargon in plain English with inline translations.

Specialized & meta (2)

• chef-assistant — Cook via technique, food science, flavor architecture.
• skill-creator — Turn a PDF / paper / methodology into a reusable skill.

Pairs with the working folder at ~/Documents/Thinking/substacker/ (inbox, corpus, shared-context, ops). Voice-profile, section-map, per-section visual identities, analogy catalog, topic ledger all live there.

Librarian (8) — corpus ingest + index

• ingest-inbox-item — Orchestrate per-file ingest: normalize → tag → score → dedupe → write seed.
• normalize-format — Convert markdown / Claude exports / transcripts / Readwise into canonical seed bodies.
• tag-by-topic — Assign 1-4 topic tags from controlled vocabulary; log pending tags.
• score-intuition-density — Compute 0-10 density from 8 explicit signals (analogy, concrete example, counterfactual, etc.).
• dedupe-against-corpus — Exact fingerprint + near-match Jaccard; link rather than merge.
• search-corpus — "What have I already thought about X?" — ranked matches with excerpts.
• update-topic-ledger — Maintain counts, last_touched, hot/warm/cold, top_seeds per topic.
• sweep-stale-seeds — Flag seeds >30d old with no links; recommend keep/promote/kill.

Intuition Builder (7) — 5 framings per topic

• generate-analogy-set — 5 framings: everyday / physical / contrarian / historical / counterfactual.
• check-analogy-novelty — Cross-check against analogy-catalog; flag reuse.
• stress-test-analogy — Find the boundary where each analogy breaks; propose a fold.
• map-analogy-to-concept — Component-by-component source → target mapping (Gentner structure-mapping).
• propose-counterfactual — "What if this weren't here?" — reveal function via subtraction.
• update-analogy-catalog — Append entry on publish; track freshness.
• voice-fitness-check — Rank framings by voice-profile analogy-direction priority (biology > organizational > sports).

Editor (10) — voice + structural review gate

• structural-review — Macro-structure: argument flow, pivot, signposting, outline.
• voice-check — Phrase-by-phrase against voice-profile don't-list.
• hedge-detector — Precision hedge (keep) vs epistemic-weakness (commit or specific).
• slop-detector — 10 signatures of AI-generated explainer slop.
• opener-critique — Confession / reframe / admission pass; news-hook / generic-opener fail.
• closer-critique — Bolded maxim / scoreboard check for series posts.
• analogy-weight-check — Mechanical-weight test; flag decorative and wrong-domain analogies.
• paragraph-rhythm-check — Long/short mix, pivot paragraph, no walls.
• citation-form-check — Papers cited with Author, Institution, Year inline.
• section-break-check — Asterisks for essays vs H2 for methodology; no mixing.

Distribution Translator (7) — platform-native rewrites

• extract-thread-spine — 5-7 point argument backbone with translatability scoring.
• hook-generator — Platform-specific first-line hook candidates.
• substack-note-rewrite — 60-180 word Note; bolded maxim closer; single link.
• x-thread-rewrite — Short / medium / long variants; ≤12 tweets; no numbering by default.
• linkedin-post-rewrite — 900-2500 chars; 210-char fold hook; 0-2 niche hashtags.
• cross-poster-blurb — 60-140 word third-person blurb for Substack cross-post feature.
• platform-voice-check — Voice audit gate with loop-back on FAIL.

Growth Analyst (8) — weekly Substack stats

• fetch-substack-stats — Primary path. Pull stats from the Substack dashboard via Claude-in-Chrome browser automation.
• ingest-substack-csv — Fallback path. Load + validate manual CSV; halt on schema drift.
• compute-baseline — Rolling 4-week median + trimmed median + IQR z-scores.
• attribute-performance — Plain-English outlier attribution with calibrated confidence.
• per-section-tracking — Per-section aggregates + prune candidates for Curator.
• fetch-public-page-stats — WebFetch supplement for stale CSV data.
• write-weekly-report — Compose 400-800 word report; hard word cap.
• update-audience-notes — Append confidence ≥ medium observations.

Trend Scout (6) — weekly ML/systems signal digest

• fetch-watchlist-sources — Pull 7-day updates from 30-50 intuition-first sources.
• summarize-signal — "Teaches X" summary; mechanism / empirical / tool / opinion / announcement / benchmark.
• cross-ref-topic-ledger — NEW / OVERLAPS seed|draft|published; reinforcement angle.
• rank-by-user-fit — Weighted-sum scoring; top 10 keeps + explicit drops.
• write-weekly-digest — Compose Friday digest; Saturday-morning readable.
• update-watchlist — Monthly: propose adds/removes via diff; writer approves.

Technical Reviewer (6) — pre-publish claim check

• claim-extractor — Extract atomic technical claims from prose.
• classify-claim — simplified-correct / simplified-boundary / wrong / contested / overclaim.
• cross-reference-claim — Find primary source (arXiv, RFC, textbook); never blog posts.
• flag-boundary-break — Fold suggestion for simplified-boundary as teaching moment.
• write-review-artifact — Compose go / go-with-hedges / no-go review.
• glossary-alignment-check — Writer's definitions vs field-standard.

Curator (9) — section-map maintainer, active + reactive

• check-corpus-readiness — 28-day + 4-post cadence gate.
• cluster-corpus-by-theme — Axial coding; Braun & Clarke thematic analysis over published corpus.
• propose-section — Cluster → named section proposal with fit-confidence.
• write-section-promise — One-sentence promise; specific, testable, non-overlapping, voiced.
• audit-drift — Flag posts no longer fitting their section's promise.
• recommend-prune — Retire / merge / reassign with reasons-to-reject.
• update-section-map — Atomic write of section-map.md with snapshot backup.
• classify-post-to-section — Assign a draft or published post to the best-fitting section.
• derive-section-voice-overlay — Draft per-section voice delta once section has ≥3 posts.

Growth Strategist (7) — quarterly zoomout

• quarterly-zoomout — 400-700 word narrative synthesizing 13 weeks of data.
• answer-uncomfortable-question — Evidence + reasoning + downside triad per question.
• section-portfolio-assessment — Healthy / drifting / candidate-for-prune per section.
• product-hiding-scan — Course / book / cohort / consulting candidates from the corpus.
• paid-tier-readiness-check — Four gates: scale / engagement / section candidate / writer capacity.
• goal-reset-proposal — Propose diff to goals.md; writer applies manually.
• identify-kill-list — Max 4 items; easiest kills first.

Clone and copy into your Claude Code skills directory:

git clone https://github.com/lyndonkl/claude.git
cd claude

# macOS / Linux
cp -r skills/* ~/.claude/skills/

# Windows
xcopy skills\* %USERPROFILE%\.claude\skills\ /E /I

skill-name/
├── SKILL.md                          # Entry point (< 300 lines)
│   ├── YAML metadata with triggers
│   ├── When to use + guardrails
│   ├── Workflow checklist
│   └── Common patterns
└── resources/
    ├── template.md                   # Ready-to-use templates (< 400 lines)
    ├── methodology.md                # Advanced techniques (< 500 lines)
    └── evaluators/
        └── rubric_[skill].json       # 10-criterion quality rubric

Progressive depth: SKILL.md for the skim, template.md for the copy-paste, methodology.md for the edge cases, rubric for self-evaluation. Every skill ships with all four.