Learn to run
process as states.

Abstract

Most financial AI products automate outputs (signals, chat, content). Statecraft automates discipline around process: explicit states, legal transitions, capacity limits, and human gates. Operators train agents by running the process; agents propose; humans accept or reject; the journal is the ground truth.

Kurogami is the broader stack: agentic assets, RWA and collateral rails, data intelligence, spatial and hardware expansion. Statecraft is the wedge that proves the thesis in production.

1. Problem

1. Discretionary skill is trapped in heads. It doesn’t transfer, rent, or compound cleanly.
2. AI agents drift without a harness: no hard states, no illegal-action rejection, no capacity law.
3. RWA and “AI + crypto” stacks often sell narrative without an operator standard.
4. Education sells content; desks need control surfaces.
5. Long-horizon assets (physical, digital, archival) lack unified process + custody + data language.

2. Thesis

Process is the asset.

If states, transitions, and gates are explicit, then:

• humans can be trained to standard,
• agents can be trained by imitation of legal behavior,
• scores can be computed (Forge Index),
• rental/sale of agents becomes possible without pretending the agent “is” the human,
• data exhaust (traces) becomes a second product.

3. Statecraft — Product

Desk Access — paid entry to the Money Process curriculum and Desk OS learning surfaces.

Symbol Strip (Control Surface)

• One row per market: state, bias, stops, legs, confidence, next legal action only
• Capacity (e.g. max concurrent risk units)
• Scenario injection + discipline drills
• Append-only journal + export
• Simulation · no brokerage · not advice

Agent Mirror

• Agent proposes next action from shared state machine
• Human Accept / Reject
• Illegal proposals fail the same as illegal human clicks

Curriculum (Money Process)

M0 foundations → states/transitions → capacity/veil → drills → review → capstone.

Agents are introduced as mirrors of process, not as yield products.

4. Architecture (conceptual)

Human operator
    ↓ logs states & decisions
Desk OS (source of truth)
    ↓ legal transition graph
Agent harness (rules, memory, evals)
    ↓ proposals only
Human gate
    ↓
Journal / Forge scores / (later) execution adapters

Hard rules: capacity can block SIGNAL; closed NEW RISK blocks new entries; out-of-sequence events do not mutate state.

5. Forge Index

MLS-style layer for process-trained agents (and later operators/assets):

• Fidelity: illegal attempt rate, capacity honors, checklist completion
• Utilization: rental demand, uptime
• Domain tags: trading process, creative process, etc.
• Not a promise of market returns

Investment angle: exposure to infrastructure that prices process quality, not only token beta.

6. Kurogami stack (expansion layers)

7. Hardware roadmap

• Phase A: Recommended desk hardware list (compute, displays, network, power).
• Phase B: “Desk appliance” — locked-down machine image running Desk OS + local model weights for proposal-only agents.
• Phase C: Edge nodes for low-latency inference; optional HSM/secure element for keys—not required for curriculum MVP.

Hardware follows software proof, not the reverse.

8. Spatial Protocol

Aim: map high-value real places and rights into navigable digital layers (twins, splat/Gaussian experiences, access-controlled parcels).

Near: known urban inventory (e.g. showcase districts), open-house style remote walkthroughs tied to clear off-chain legal title.

Later: integration with remote sensing / orbital data providers for monitoring and archival—data licenses and products, not claims of sovereign ownership of space.

Spatial is a distribution and documentation surface for RWA and events—not a replacement for property law.

9. RWA and long-horizon preservation

Standard RWA path: tokenize or document claims where regulation and custodians allow (real estate, commodities, credit, art/collectibles)—always with legal wrappers.

Preservation vertical (careful framing)

Institutions and collectors need audit trails, custody, and access control for assets and datasets that outlive hype cycles—including cultural property, scientific archives, and lawful anomaly / aerial phenomena open-data projects. Kurogami’s contribution is process + ledger + agentized monitoring, not unverifiable cosmological claims.

“Extraterrestrial preservation” in this whitepaper means: long-horizon preservation of rare observational data and related IP under clear chain-of-custody, if and when partners and law allow—not retail “space deed” fantasies.

10. Token / NFT posture (high level)

• Desk Access can remain a pure SaaS/education SKU.
• Optional later: membership NFTs as access keys, agent NFTs as licenses to run a scored policy, creator brands tied to agent performance—with securities/consumer-law review before launch.
• No roadmap dependency on a single meme ticker.

11. GTA / virtual economy (optional bridge)

Virtual worlds are distribution: verified access, digital twins, creator mods, events. Real value stays in process agents, data, and legal RWA rails. Virtual layers amplify attention; they do not substitute custody.

12. Business model (illustrative)

• Desk Access (high-ticket education + software surfaces)
• Seat expansions / team desks
• Later: agent listing fees, rental take-rate, data licenses, hardware SKUs, institutional process audits
• RWA and preservation: success/structuring fees with licensed partners

Projections belong in a separate numbers deck with disclaimers; this whitepaper states mechanics, not guaranteed IRRs.

13. Competition

14. Risks

Execution complexity; regulation of education vs advisory vs securities; agent misuse if gates are removed; market risk if any live trading adapters ship; narrative overreach on spatial/UAP themes; key-person dependence until process is fully documented.

Mitigations: simulation-first, human gates, clear disclaimers, phased licensing, journal-based evals, narrow v1 scope.

15. Governance / build ethos

• Self-funded bias; paid seats fund depth
• Hierarchy of contribution over social volume
• Kurogami backend, Statecraft frontend until proof is undeniable