Worked Examples

Phygital Loop

Three agent types, four instrument families, two feedback instruments, typed decisions at arbitration points.

Reading Example 1

Code	Reads As
HA[L3]	Human Agent, authority level 3 (can override any gate)
DA[L1]	Digital Agent, authority level 1 (executes only)
ST-R01	Settlement Token Recorder #1 (logs, no control)
GT-C01	Governance Token Controller #1 (adjusts continuously)
IT-G01	Incentive Token Gate #1 (token required to proceed)
VFL-DX01	VFL Exclusive Decision #1 (one path chosen)
VV-G01	Verifiable Value Gate #1 (pass/fail on value criteria)
LR-C01	Learning Controller #1 (adjusts agent parameters)
[P]	Connection flows on Pass
[F]	Connection flows on Fail
[E:receipt]	Connection fires on receipt event
[T:cycle]	Connection fires on time cycle

Three Decision Points

The diagram exposes three critical choice nodes:

1. Signals agree? (◇X) — Do physical and digital instruments produce consistent evidence?
2. Who arbitrates? (◇X) — Protocol rules, human override, or market mechanism?
3. Value sufficient? (VV-G01) — Does the outcome meet the gate threshold?

---

Human + Orchestrator

One human, one orchestrator agent (Nav as priorities-nav.md would express it), local bodies (laptop, phone), a DePIN sensor on the shared substrate (e.g. GEODNET RTK), and a verifier agent that only passes receipts when evidence clears the gate. Same symbol library as example 1 — no new primitives.

Reading Example 2

Code	Reads as
HA[L3]	Human Agent at L3 — can override any gate
OA[L2]	Nav as Orchestrator Agent at L2 — routes within scope
PA[L1]	Physical Agent at L1 — laptop, phone, DePIN sensor execute only
DA[L2]	Digital Verifier Agent at L2 — decides whether outputs pass
TS-T01	Universal Machine Time Sync Transmitter — anchors readings (see Economy of Things)
VV-G01	Verifiable Value Gate — receipt only if value clears threshold
ID-G01	Identity Gate — each body proves identity before participating
learning loop arc	Feedback crosses back to the human — same diagram, tighter behaviour next cycle

The Key Pattern

Today the human owns the laptop and phone bodies directly. Tomorrow the same drawing absorbs more DePIN-owned instruments under the same ID-G01 + VV-G01 contract.

The topology does not change — the substrate underneath grows.

That is publishing the monomyth in operating form, not in prose.

---

Questions

Can you read a diagram built by someone in a different discipline and correctly execute it?

• In Example 1, what happens if physical and digital instruments disagree? Which decision node (◇X or ◇X) catches that — and can a protocol-level rule ever override human authority HA[L3]?
• In Example 2, the learning loop crosses back to the human. How would you encode "stop learning if the human has been overridden twice in a row"?
• If you needed to add a fourth agent type (Collective Agent?) or a new connection type (consensus flow?), where would you add it without breaking the existing notation?

---

Next: See Context — related concepts and connections.