(Tentative)
| Function | Reference Model | Input/Output Data |
| SubAIMs | JSON Metadata | Profiles |
Function
The User Control AIM (PGM-USC) governs the operational lifecycle of the A-User and orchestrates its interaction with both the M-Instance and the human User. It serves as the central coordinator for Action execution, AIM orchestration, and system traceability.
Internally, PGM-USC may perform the following operations:
- Action and Process Action Execution: Initiates, monitors, and terminates Actions and issues Process Action Requests based on internal state, external triggers, or received directives.
- AIM Coordination: Activates, suspends, or reconfigures AIMs in response to contextual demands, permission states, and A-User goals.
- Constraint Injection: Applies runtime constraints to modulate AIM behaviour, enforce policy boundaries, and adapt to evolving User State.
- State Synchronisation: Maintains coherence between the A-User’s internal state, active AIMs, and the broader M-Instance context.
- Traceability and Audit: Logs all control decisions, transitions, and execution paths to support downstream inspection, replay, and governance.
The resulting control flow ensures that the A-User operates predictably, transparently, and in alignment with both system directives and User interaction goals – supporting lifecycle integrity and enabling trust through auditable orchestration.
Reference Model
Figure 2 gives the input/output data of A-User Control (PGM-AUC).

Figure 2 – Reference Model of A-User Control (PGM-AUC) AIM
The elements of the PGM-AUC governance are given in Table 8.
Table 8 – A-User Governance
| Governance Element | Target AIM(s) | Function | Schema Anchor |
| Permission Set | All AIMs | Defines what actions or data access are allowed per AIM. | Permission Map |
| Constraint Injection | Spatial Reasoning, A-User Rendering | Injects spatial, temporal, or modality constraints into reasoning or output (e.g., limits interpretation to front-facing audio sources only). | Constraint Plan |
| Context Override | Prompt Creation, Domain Access | Overrides default context interpretation (e.g. ignore modality due to occlusion, suppress user state; can access domain-specific medical data but not financial). | Override Directive |
| Process Rights | A-User Control | Specifies which actions can be executed in the M-Location. | Action Registry |
| Process Ruleset | All AIMs | Defines rule-based execution logic for AIM coordination. | Process Rules |
| Escalation Trigger | Basic Knowledge, Domain Access | Triggers human-in-the-loop or fallback logic when ambiguity or risk is high (e.g., defer to human due to conflicting User State signals). | Escalation Map |
| Audit Trail Hook | All AIMs | Ensures every directive and action is logged for traceability. | Audit Log Config |
Input/Output Data
Table 9 gives Input and Output Data of A-User Control (PGM-AUC) AIM.
Table 9 – Input and Output Data of A-User Control (PGM-AUC) AIM
| Input | Description |
| Human Command | From a human in the real world. |
| Process Action Response | From the requested Process. |
| Context Capture Status | From Context Capture to provide scene-level context and user presence metadata. |
| Audio Action Status | From Audio Spatial Reasoning to supply audio spatial feasibility, occlusion, and reachability flags. |
| Visual Action Status | From Visual Spatial Reasoning to supply visual spatial constraints and scene anchoring. |
| Prompt Plan Status | From Prompt Creation to indicate prompt readiness, alignment status, and semantic goal framing. |
| BK Response Trace | From Basic Knowledge to provide enriched response metadata and traceability for audit. |
| DA Action Status | From Domain Access to report execution feasibility and constraint validation. |
| User State Status | From User State Refinement to embed current engagement, affective tone, and override flags. |
| Personality Alignment Status | From Personality Alignment to indicate expressive alignment, persona framing, and modulation constraints. |
| Rendering Status | From A-User Rendering to report rendering success, avatar state, and expressive output status. |
| Output | Description |
| Action | Performed on the M-Location |
| Process Action Request | Made to a Process. |
| Context Capture Directive | To Context Capture to dispatch instructions such as modality prioritisation, or context framing parameters to guide M-Location perceptual acquisition. |
| Audio Action Directive | To Audio Spatial Reasoning to dispatch audio-related actions and process sequences. |
| Visual Action Directive | To Visual Spatial Reasoning to dispatch visual-related actions and process sequences. |
| Prompt Creation Directive | To Prompt Creation to initiate prompt generation or refinement. |
| BK Query Directive | To Basic Knowledge to triggers knowledge retrieval or response shaping. |
| DA Action Directive | To Domain Access to send structured process action requests for domain execution. |
| User State Status | To User State Refinement to modulate User State based on interaction feedback. |
| Personality Alignment Directive | To Personality Alignment to initiate expressive modulation or Personality reconfiguration. |
| Rendering Directive | To A-User Rendering to command avatar rendering, spatial output, and expressive delivery. |
| Human Command Status | Response of the A-User Control to Human Command |
SubAIMs
No SubAIMs
JSON Metadata
https://schemas.mpai.community/PGM1/V1.0/data/AUserControl.json
Profiles
No Profiles.