Start with authorization, not indicators
M.A.I.C. is a crypto-native capital operating environment. The safest reading order is posture → route → risk → authorization → evidence/replay. Analytical evidence may explain a market condition, but it does not automatically grant capital permission.
Favorable evidence can exist while deployment remains constrained, denied, or unchanged.
Portfolio structure can remain visible under STANDBY. A displayed allocation envelope is not an instruction to create a new position.
How decision authority flows
M.A.I.C. is not reducible to one score, one predictive model, or a simple if-then rule chain. Distinct analytical surfaces observe different market dimensions; authority is determined only after conflict, prior state, and governance constraints are resolved.
Access classes expand authority, not merely panel count
Class 01 observes public state. Class 02 governs the confirmed portfolio route. Class 03 opens earlier/deeper protected intelligence and interactive rationale. Class 04 uses the same intelligence ceiling as Class 03 for the future institutional mandate layer.
| Access | Job | Available | Still protected / pending |
|---|---|---|---|
| Class 01 · Public Research | Observe the market | Public weekly-confirmed posture, public evidence, masked topology, public historical evidence | Interactive portfolio/rationale controls |
| Class 02 · Tactical Portfolio | Govern the portfolio | Confirmed weekly portfolio governance, BTC-first allocation buckets, risk review, scenario, deterministic rationale, bounded historical review | Protected inference, deep archive, interactive Operating Rationale |
| Class 03 · Allocation Intelligence | See deeper / earlier | Class 02 + protected inference, topology depth, deep reconstruction, projection memory, Operating Rationale, reporting | System asset classification appears only when supplied by the active governed model |
| Class 04 · Institutional Mandate | Govern a mandate | Same market-intelligence ceiling as Class 03 | Mandate scope does not increase the market-intelligence ceiling |
Evidence discipline: 4H confirmed evidence
Source telemetry may move at its native cadence. M.A.I.C. synchronizes decision evidence on four-hour boundaries to suppress intraday noise and unnecessary posture churn. This is a capital operating environment, not an HFT engine.
The public directive is a separate weekly-confirmed governance state. A fresh 4H evidence boundary can update what the system observes without automatically changing what the public portfolio is authorized to do.
Seven Terminal phases
Each phase answers one governance question. Navigation never expands authority.
| Phase | Surface | Governance question | Interpretation boundary |
|---|---|---|---|
| 01 | Operational Posture | What can the portfolio do now? | Posture is the primary governed state. |
| 02 | Liquidity Governance | Where is money moving? | Flow evidence can support or challenge a route; it does not open one by itself. |
| 03 | Survivability Corridor | What can hurt the portfolio? | Risk boundaries can constrain otherwise favorable evidence. |
| 04 | Exposure Authorization | What can actually be deployed? | STRUCTURAL ALLOCATION ≠ ACTION AUTHORIZATION. |
| 05 | Topology Governance | What deserves capital? | RANKING ≠ PERMISSION. Candidate evidence is not automatic bucket membership. |
| 06 | Scenario Governance | What if the market moves differently? | SCENARIO ≠ AUTHORIZATION. Probability is supporting evidence. |
| 07 | Governance Evidence | Did the system behave as intended? | REPLAY ≠ LIVE AUTHORIZATION. |
Canonical directive semantics
Signal color is a canonical state cue. The operational meaning of the directive is more important than the color itself.
| Directive | Canonical color | Operational meaning |
|---|---|---|
| OFFENSE | Blue | Broad crypto-risk deployment is open within governed model constraints. |
| CORE | Green | BTC-first / bluechip model exposure remains prioritized while broader risk stays controlled. |
| STANDBY | Yellow | No new deployment. Existing model structure may remain visible; this is not automatic liquidation. |
| DEFENSE | Red | Higher-beta risk-reduction state: model exposure compresses toward stronger bluechip/cash structure. |
| DEGRADED / UNAVAILABLE | Grey | Normal state authority is constrained because evidence is incomplete or unavailable. |
OFFENSE, CORE, STANDBY, and DEFENSE are human-facing state names. They do not change backend machine contracts, historical state logic, or API semantics.
Historical Market Replay & Bottom Projection
Historical Market Replay reconstructs the governed context of a selected epoch. It must not redefine the live directive or leak a replay-specific annotation into the current public posture.
Inside Historical Market Replay only, the first full STANDBY week immediately after a governed DEFENSE may retain a red epoch transition marker so the replay can preserve the same top-cycle / bottom-projection episode. The public state remains STANDBY and its operational meaning remains “no new deployment.”
Bottom Projection is therefore interpreted as an episode state machine, not as a color rule attached only to ROTATE. The interpreter reads every canonical posture to determine whether an episode is opened, observed, cancelled, or frozen.
| Observed posture | Replay / projection interpretation |
|---|---|
| DEFENSE | Opens the candidate top-cycle / bottom-projection episode. |
| STANDBY | Preserves the no-new-deployment state; the first post-DEFENSE replay epoch may keep red lineage solely to show episode continuity. |
| CORE | If it appears before projection freeze, the pending episode is cancelled as an invalid / false top candidate. |
| OFFENSE | If it appears before projection freeze, the pending episode is cancelled as an invalid / false top candidate. |
Projection freeze remains a governed weekly observation condition. The projected range is probabilistic downside evidence, not a guaranteed target, not a live sell order, and not a replacement for the current posture.
Historical Review vs Temporal Posture Memory
Historical Governance Review is bounded evidence review for Class 02+. Temporal Posture Memory is Class 03+ arbitrary-epoch selection, deep reconstruction, and live/archive restore. Lower classes may discover the capability without receiving its controls.
Decision Rationale vs Operating Rationale
Class 02 receives deterministic Decision Rationale derived from confirmed state. Class 03 receives Operating Rationale: ASK THE SYSTEM WHY. It may interrogate evidence, conflict, uncertainty, history, constraints, and change conditions while remaining tethered to the governed decision.
Routing Access, command palette, and Keyboard Mode
Commands and shortcuts accelerate navigation and workflow. They delegate to canonical controls; there is no shortcut around access, posture, survivability, or routing denial.
Keyboard shortcuts
| Key | Route / action |
|---|---|
| 1-7 | Phase navigation |
| O / G | Observation / Governance lens |
| B / I | Baseline / Inference visibility |
| W / V | Workflow / Supervision |
| C / S / E | Command / System / Evidence depth |
| R | Operating Rationale (only when authorized) |
| L | English / Indonesia |
| ? or / | Shortcut reference |
| Ctrl+K | Command Search / Routing Access |
| Ctrl+M | Survivability Review |
| Ctrl+Shift+K | Keyboard Mode on/off |
Command Search routes
| Command | Minimum access | Destination |
|---|---|---|
| /posture | Class 01 | Operational Posture |
| /risk | Class 02 | Survivability Corridor |
| /allocation | Class 02 | Exposure Authorization |
| /deploy | Class 02 | Confirmed Deployment Review |
| /topology | Class 03 | Topology Governance / protected depth |
| /audit | Class 02 | Historical Governance Review |
| /archive | Class 03 | Temporal Posture Memory reconstruction |
| /supervision | Class 03 | Allocator Oversight Surface |
| /report | Class 03 | Governance report |
| /rationale | Class 03 | Operating Rationale |
Review Lens, Workspace, and Evidence Depth
Review Lens changes observation/governance framing; Workflow vs Supervision changes operating surface; Command/System/Evidence changes disclosure depth. These are presentation controls, not decision authority.
Evidence freshness and degraded state
The Terminal distinguishes fresh governed evidence from stale, incomplete, or unavailable telemetry. When source evidence cannot support normal confidence, the surface should say so explicitly rather than silently extending a prior state.
Use the displayed evidence boundary and next governed refresh as the primary freshness reference. A degraded label describes evidence quality; it does not create a new market directive.
Mobile controls
Phones use one persistent rail: POSTURE / LIQUIDITY / RISK / MORE. MORE routes Exposure, Topology, Scenario, Historical Review, Memory, Commands, and Evidence according to class. CONTROLS and WHY are contextual utilities. Wide analytical canvases pan or scroll inside bounded containers rather than overflowing the document.
Security, privacy, and authority boundary
Protected access is enforced by the authorized service boundary, not by what happens to be visible in the browser interface. Presentation controls cannot manufacture a higher access class. Physical execution, custody, brokerage, and capital transfer remain outside the M.A.I.C. Terminal.
Troubleshooting
If state appears stale, check the evidence boundary, next governed refresh, and degraded status before forcing reload. If a control is visible but unavailable, read its access or capability label: discoverability does not guarantee interaction authority. If replay appears different from live state, confirm whether you are viewing an epoch annotation or the current governed posture.
Short glossary
CONFIRMED = canonical weekly governance state. EARLY / INFERENCE = protected provisional evidence. GOVERNED = state after constraints and authority are applied. OBSERVATION = evidence that does not automatically become permission. AUTHORIZATION = capital consequence allowed by the system. STRUCTURAL ALLOCATION = portfolio envelope/reference, not automatic deployment. RANKING = candidate ordering, not portfolio permission. SCENARIO = probabilistic supporting evidence, not authorization. REPLAY = historical context, not live authority. DEGRADATION = evidence is not fresh or complete enough for normal confidence.