Highlights
43 new skills across 3 tiers, 1 new domain (research/), and 5 new C-suite advisors.
Skill count: 294 → 338 (+44 incl. previous)
Domain count: 15 → 16
Python tools: 720+ → 784+ (stdlib only)
Tier 1 — 17 deep skills (engineering, commercial, compliance, marketing)
Engineering (6):
feature-flags-architect— flag audit, rollout simulator, kill switch runbookchaos-engineering— experiment designer, blast radius calculator, gameday plannerkubernetes-operator— CRD validator, operator scaffold, reconciliation auditazure-cloud-architect— architecture validator, cost estimator, WAF scorergcp-cloud-architect— architecture validator, cost estimator, CAF scorerdata-quality-auditor— DQ check runner, schema drift detector, freshness monitor
Business-growth — commercial mechanics (4):
deal-desk·channel-economics·partnerships-architect·commercial-policy
RA/QM — audit-prep playbooks (6, new ra-qm-team/audit-prep/ subfolder):
soc2-audit-prep·gdpr-audit-prep·fda-qsr-audit-prep·ai-act-readiness·aims-audit·compliance-readiness
Marketing (1):
aeo— Answer Engine Optimization (LLM citation tracking; distinct from AI-SEO)
Tier 2 — 14 deep skills (C-suite, product, new research/ domain)
C-level advisory (5):
chief-ai-officer-advisor— AI maturity, investment planner, risk register (NIST AI RMF / ISO 42001 / EU AI Act)chief-data-officer-advisor— data maturity, DAMA-DMBOK governance audit, platform evaluatorchief-customer-officer-advisor— CX maturity, churn intervention, VoC program designergeneral-counsel-advisor— legal risk register, contract portfolio analyzer, regulatory calendarvpe-advisor— engineering org health, DORA/DevEx dashboard, capacity planner
Product team (5):
product-analytics— metric tree designer, event taxonomy auditor, retention cohort analyzerapple-hig-expert— HIG compliance checker, component pattern lookup, accessibility auditorresearch-summarizer— synthesis organizer, insight quality scorer, findings brief generatorspec-to-repo— PRD-to-tickets decomposer, branch naming validator, PR scope analyzerroadmap-communicator— audience translator, confidence band generator, diff reporter
Research — NEW DOMAIN (4):
litreview— PRISMA-aligned search strategy, source quality scoring, thematic synthesisgrants— funder fit, proposal structure (NIH/NSF/SBIR/foundation/DARPA), budget realismpatent— prior-art search planner, claim landscape mapper, patentability scorerdossier— outline generator, source triangulation (Admiralty Code), fact/inference separator
Tier 3 — 12 PM skills (PM 54 → 66)
Strategy frameworks — new project-management/strategy-frameworks/ subfolder (5):
business-model-canvas— Osterwalder 9-block + cross-block coherence validatorlean-canvas— Ash Maurya startup canvas + unfair-advantage disciplineswot-analysis— SWOT + TOWS matrix + evidence auditporters-five-forces— industry analysis + strategy translationansoff-matrix— growth quadrants + stage-appropriate investment mix
Go-to-market — new project-management/gtm/ subfolder (2):
gtm-strategy— ICP × motion × channels × messaging × T-90 to T+90 sequenceideal-customer-profile— 8-dimension ICP + qualification rubric + scoring
Discovery additions (2):
opportunity-solution-tree— Teresa Torres OST: outcome / opportunity / solution / testmetrics-dashboard— NS + inputs + guardrails + operational; anti-vanity audit
Execution additions (3):
stakeholder-map— Power × Interest 2x2 + DACI + blocker conversion planstest-scenarios— 7-category coverage (happy / edge / error / empty / concurrent / a11y / security)sprint-plan— capacity math + commit/stretch discipline + DoD audit
Documentation
- README: added Claude Cowork plugin install section (3-step flow — Customize → Browse plugins → repo name)
- bundles.json: 15 → 20 bundles. New:
research-stack,fractional-c-suite,customer-experience-stack,strategy-frameworks-kit,gtm-launch-kit - Per-domain CLAUDE.md refreshed with new skill counts and Tier 2/3 additions
- docs/SKILLS.md updated with Tier 2 and Tier 3 sections
Stats
- 285+ new files
- ~73K lines (markdown + Python)
- 110 stdlib-only Python tools (no external deps)
- All scripts pass
--helpsmoke test - All bundles validate; all skill paths resolve