00 — Building Blocks Index (MOC)
What this vault is
A living documentation of the operating system behind the business: the strategic architecture (how the business is structured and grows) and the agentic/automation system (how that growth gets executed by agents). It exists so that learning-while-building never gets lost — every model, test, and decision has a home.
This is a Map of Content. Start at 01 - System Overview, then dive into whichever layer you need.
The two halves of the system
STRATEGY (what & why) AUTOMATION (how)
┌─────────────────────────────┐ ┌──────────────────────────────┐
│ Three Entities │ │ BAU vs Pilot Framework │
│ ├─ Market (Brand/Prod/Dist)│ │ Automation Decision Matrix │
│ ├─ Capital │ → │ Multi-Agent Orchestration │
│ └─ Operation │ │ Agent Operating Files │
│ Governing Artifacts ───────┼───┤ (policies / skills / guard) │
└─────────────────────────────┘ └──────────────────────────────┘
│ │
└──────────► Driver Tree / KPIs ◄──────────┘
(the shared measurement spine)
Strategy layer
-
Three Entities — Market · Capital · Operation (the three top-level entities)
-
Market — own the demand; umbrella over the three pillars below
- Branding — moats · advantage · power
- Product — value architecture · JTBD · value proposition · demand
- Distribution — the Growth pillar (acquisition → conversion → retention)
- Acquisition Channels — paid · content · virality · social commerce
- Conversion and Engagement — website + social media
- Retention — stickiness · database · community
-
Capital — bookkeeping, accounting, finance; raise · allocate · wield
-
Operation — costing, supply chain, inventory, the fulfilment spine
-
Governing Artifacts — the context layer that powers both strategy and agents
-
Static and Dynamic Functions - prioritization lens: stable reusable functions first, market-sensitive functions second
Tactical layer
- Tactical — the how: concrete plays beneath Strategy, split by maturity
- Practical — built & executed → documented, trusted
- Theoretical — observed & collected → brainstormed, unproven until built
Automation layer
- Part B2 - BAU vs Pilot Framework — deciding what to automate vs keep human
- Part B1 - Automation Decision Matrix — the practical 2×2 + scoring
- Part A2 - Multi-Agent Orchestration — reference architecture for agent pipelines
- Part C1 - Case Study - Peptide TikTok Network — the worked example, end to end
- Part A1 - Agent Operating Files — policies.md · skills.md · guardrail.md and how they fit
Critique & corrections
- Critique and Best Practices — where the model is strong, where to adjust
- Risks and Open Questions — what could break, what’s still undecided
Templates (copy these)
- Agent operating files: claude.template · policies.template · skills.template · guardrail.template · eval.template · changelog.template
- Workflow: content-brief.template · agent-spec.template · decision-log.template
The single most important habit
The pain that started this vault — “I forgot what I tested and have to relearn” — is solved by one artifact: a Experiment Log. Treat it as a lab notebook. Every test, every learning, dated and tagged. See The decision log is your highest-leverage artifact.