(Policy-grade · AI-ready · v0)
This is the minimum executable form of Civilisation OS.
If you follow this page exactly, you can diagnose, intervene, and verify repair without ideology or overreach.
0) The One Rule
Never run more than ONE Recovery Mode at a time.
Everything else flows from this.
1) Define the System Boundary (S)
Answer only these:
- What system is failing? (student / school / company / city / nation)
- What is included?
- What is explicitly excluded?
If you can’t draw the boundary, you can’t repair the system.
2) Score DLR by OS (0–5 each)
Score each OS separately:
| OS | D (Depth) | L (Load) | R (Repairability) |
|---|---|---|---|
| Education OS | |||
| Governance OS | |||
| Production OS | |||
| Constraint OS |
Rule:
- The lowest R usually defines the failure.
- High L + Low R = collapse acceleration risk.
3) Identify the Dominant Collapse Signature
Choose one best match:
- Capability Stall
- Incentive Inversion (Goodhart)
- Institutional Hollowing
- Maintenance Debt Spiral
- Fragility Cascade
- Scarcity Bind
- Late Correction / Irreversibility
If more than one fits, pick the earliest in the chain.
4) Select Exactly ONE Recovery Mode
| Signature | Recovery Mode |
|---|---|
| Capability Stall | RM-01 Capability Rebuild |
| Incentive Inversion | RM-06 Truth Restoration |
| Institutional Hollowing | RM-06 Truth Restoration |
| Maintenance Debt Spiral | RM-11 Maintenance & Renewal |
| Fragility Cascade | RM-10 Resilience & Redundancy |
| Scarcity Bind | RM-13 Load Reduction |
| Late Correction | RM-00 Stabilise the Loop |
Do NOT improvise.
Wrong mode = faster collapse.
5) Compile OSME-e/t (Execution Plan)
O — Outcome
What must improve in reality (measurable)?
S — System Boundary
Reconfirm scope (no expansion).
M — Mechanism
Why the failure exists (one paragraph).
E — Execution
3–6 concrete actions only.
e/t — Energy over Time
- Can this be sustained?
- For how long?
If energy > capacity → stop and rescope.
6) Define Probes & Retests (Falsifiability)
Minimum 3 probes:
- 1 short-horizon (weeks)
- 1 mid-horizon (months)
- 1 long-horizon (structural)
Each probe must have:
- a clear metric
- a pass/fail threshold
- a retest date
No probes = no claims.
7) Retest → Decide
At retest:
- Pass: continue same mode
- Mixed: refine execution
- Fail: re-diagnose DLR, choose a new single mode
Never stack modes without retesting.
8) Safety Stops (Hard Rules)
Immediately halt execution if:
- truth channels are suppressed
- probes are gamed
- scope quietly expands
- coercion replaces consent
- metrics improve but reality worsens
These indicate Governance OS failure.
9) When to Stop
Stop when:
- R ≥ 3 across critical OS
- collapse signatures fade
- probes stabilise across two retest cycles
Then transition to maintenance, not optimisation.
Mental Model (Remember This)
Education builds capacity
Governance steers behaviour
Production converts capacity into power
Constraints decide the final outcome
Collapse happens when repairability falls below system speed.