Function
Ref. Model
I/O Data
SubAIMs
JSON MData
Profiles
Ref. Software
Conformance
Performance
1 Functions
The A‑User Control (PGM‑AUC) AIM serves as the central coordinator for Action execution, AIM orchestration, and system traceability. It governs the lifecycle of the A‑User and orchestrates its interactions with the responsible human, the M‑Instance, and the Processes and Items with which the A‑User interacts.
A‑User Control holds goal authority but not necessarily epistemic capability: it carries the objectives given to it by the responsible human or by specification, and it asserts those objectives authoritatively. It holds the World Model but it does not itself understand the environment or the User. That understanding resides in the reasoning AIMs. Consequently, a Directive is binding as to the goal it sets and the constraints it imposes, but it is advisory as to the conclusion: an AIM that reasons decides its own conclusion, and an AIM that executes attempts the goal subject to the conditions of the world. In neither case may an AIM decline the goal, but any AIM may report that the goal was not achieved, or was achieved differently, and why.
Accordingly, a Status is not merely a record of completion but a decision-and-outcome report. Every Status states the goal the AIM understood, the outcome reached, and a reason. Executor AIMs (such as Context Capture and Context Enhancement) report the outcome and, where a goal could not be met, the world condition that prevented it (for example, occlusion or unreachability). Reasoning AIMs (such as User State Refinement, Basic Knowledge, and Personality Alignment) report the conclusion they reached and, where A‑User Control proposed a correction, whether they adopted, partially adopted, or declined it, and why.
Therefore, A‑User Control does not impose a fixed schedule on the AIMs it cannot direct in detail; it sets goals and constraints, lets each AIM work, inspects the reported outcome, and may call again with a revised goal or a proposed correction. Every call is answered, and the initiative to call always rests with A‑User Control. In this way, A‑User Control bounds and steers the A‑User’s behaviour by exception rather than by micromanagement, judging each AIM’s report against the objective it was given rather than against a correctness it is not equipped to assess.
A‑User Control also governs all access to A‑User Storage. Because routing every read and write through A‑User Control would make it a bottleneck, A‑User Control is never in the storage data path; the orchestrated AIMs perform their own storage operations. A‑User Control governs that access in two complementary ways: it instructs each producing AIM, through that AIM’s Directive, what to store and what it may read; and it configures A‑User Storage, through the AUS Directive, with the session’s storage scopes, retention and life cycle policy, and the access rights A‑User Storage enforces. Writes are performed only as instructed; session‑scope reads are performed by an AIM as it needs them, while historical‑scope reads are permitted only under an authorisation A‑User Control establishes and A‑User Storage enforces.
A‑User Control sends Directive messages to AIMs to implement Instructions within the Rights the A‑User holds and the Rules applicable to the M‑Location, and tracks execution of Directives using Status messages received from A‑User AIMs.
The resulting control flow ensures that the A‑User operates predictably, transparently, and in alignment with human Commands, supporting lifecycle integrity and enabling trust through auditable orchestration.
2 Reference Model
Figure 1 depicts the Reference Model of the A‑User Control (PGM‑AUC) AIM.
Figure 1 – Reference Model of the A‑User Control (PGM‑AUC) AIM
The A‑User Control AIM exercises its activity by implementing one of eight Instructions:
- Perception and Environment Capture (PEC): Configure perceptual subsystems to sense the responsible human in the Universe, the User in the M‑Instance, or the contents of the relevant M‑Location.
- Goal and Language Acquisition (GLA): Capture and segment multimodal expressions of the responsible human and/or the User. This corresponds to giving identity and spatial relationship to the objects in the scene and to understanding the User.
- Prompting and Knowledge Query (PKQ): Enable structured contextual representation and semantic grounding of perceptual and state information. The A‑User is now in a position to make sense of what it perceives and interprets.
- Goal and Intent Interpretation (GII): Based upon the results of the previous Instruction, trigger deliberative processing to determine the appropriate communicative behaviour of the A‑User. The A‑User Control is now able to decide the stance it should take.
- Policy, Rights, and Feasibility (PRF): Validate intended behaviour with respect to governance Rules, User Entity State constraints, human commands, and domain feasibility conditions. This includes establishing the A‑User Storage access rights enforced for the session. The A‑User Control knows that the A‑User actions must comply with a variety of constraints.
- Plan Construction and Execution (PCE): Orchestrate execution of the behaviour based on Statuses reported by the AIMs, including speech and actions. The A‑User Control is now able to execute the actions after taking constraints into account.
- Conflict Management and Escalation (CME): Detect unresolved inconsistencies or conflicts and escalate to the responsible human when required. Various impediments may be encountered and modifications to the action decided.
- Avatar Formation and Rendering (AFR): Enable synthesis and rendering of the speaking avatar. The A‑User’s Avatar is formed and can be rendered.
3 I/O Data
The A‑User Control (PGM‑AUC) AIM exchanges typed Directive and Status messages with each of the A‑User AIMs it orchestrates. Each AIM receives a typed Directive and returns a typed Status to A‑User Control.
Table 1 gives the Input and Output Data of the A‑User Control (PGM‑AUC) AIM.
| Input | Description |
|---|---|
| Human Command | Command from the responsible human in the real world. |
| Process Action Response | From a Process that has received a Process Action Request. |
| CXC Status | Scene‑level context and User presence reported by Context Capture. |
| CXE Status | Spatial feasibility, occlusion, reachability flags reported by Context Enhancement. |
| PRC Status | Prompt readiness, alignment status, semantic goal framing reported by Prompt Creation. |
| BKN Status | Enriched response metadata and traceability reported by Basic Knowledge. |
| DAC Status | Execution feasibility and constraint validation reported by Domain Access. |
| USR Status | Current engagement, affective tone, override flags reported by User State Refinement. |
| PAL Status | Expressive alignment, persona framing, modulation constraints reported by Personality Alignment. |
| AUF Status | Avatar formation success, avatar state, expressive output status reported by A‑User Formation. |
| AUS Status | Storage state, per‑operation outcomes, scope and retention enforcement, integrity and capacity, and lifecycle outcomes reported by A‑User Storage. |
| Output | Description |
| Action | Action performed by A‑User on the M‑Instance. |
| Process Action Request | Request made by A‑User to an M‑Instance Process. |
| CXC Directive | Instructions for perceptual acquisition sent to Context Capture. |
| CXE Directive | Context Enhancement‑related actions and sequences sent to Context Enhancement. |
| PRC Directive | Prompt generation or refinement instructions sent to Prompt Creation. |
| BKN Directive | Request for knowledge retrieval or response shaping sent to Basic Knowledge. |
| DAC Directive | Request for domain execution sent to Domain Access. |
| USR Directive | Request to modulate User State based on interaction feedback sent to User State Refinement. |
| PAL Directive | Request for expressive modulation or Personality reconfiguration sent to Personality Alignment. |
| AUF Directive | Request for avatar formation, spatial output, and expressive delivery sent to A‑User Formation. |
| AUS Directive | Instructions establishing storage scopes, retention and lifecycle policy, session identification, and the access rights A‑User Storage enforces, sent to A‑User Storage. |
| Human Command Status | A‑User Control response to Human Command. |
4 SubAIMs
No SubAIMs.
5 JSON Metadata
https://schemas.mpai.community/PGM1/V1.0/AIMs/AUserControl.json
6 Profiles
No Profiles.
7 Reference Software
Not part of this specification.
8 Conformance Testing
Table 2 provides the Conformance Testing Method for the A‑User Control (PGM‑AUC) AIM.
If a schema contains references to other schemas, conformance of data for the primary schema implies that any data referencing a secondary schema shall also validate against the relevant schema, if present, and conform with the Qualifier, if present.
| Receives | Human Command | Shall validate against Human Command schema. |
| Process Action Response | Shall validate against Process Action schema. | |
| CXC Status | Shall validate against CXC Status schema. | |
| CXE Status | Shall validate against CXE Status schema. | |
| PRC Status | Shall validate against PRC Status schema. | |
| BKN Status | Shall validate against BKN Status schema. | |
| DAC Status | Shall validate against DAC Status schema. | |
| USR Status | Shall validate against USR Status schema. | |
| PAL Status | Shall validate against PAL Status schema. | |
| AUF Status | Shall validate against AUF Status schema. | |
| AUS Status | Shall validate against AUS Status schema. | |
| Produces | Action | Shall validate against Action schema. |
| Process Action Request | Shall validate against Process Action schema. | |
| CXC Directive | Shall validate against CXC Directive schema. | |
| CXE Directive | Shall validate against CXE Directive schema. | |
| PRC Directive | Shall validate against PRC Directive schema. | |
| BKN Directive | Shall validate against BKN Directive schema. | |
| DAC Directive | Shall validate against DAC Directive schema. | |
| USR Directive | Shall validate against USR Directive schema. | |
| PAL Directive | Shall validate against PAL Directive schema. | |
| AUF Directive | Shall validate against AUF Directive schema. | |
| AUS Directive | Shall validate against AUS Directive schema. | |
| Human Command Status | Shall validate against Human Command Status schema. |
9 Performance Assessment
Not part of this specification.
