Go to PGM-AUA V1.0 AI Modules

Function
Ref. Model
I/O Data
SubAIMs
JSON MData
Profiles
Ref. Software
Conformance
Performance

1 Functions

The Prompt Creation (PGM‑PRC) AIM assembles, organises, and submits a structured representation of the current interaction situation of the A‑User vis‑à‑vis the Space and the User to Basic Knowledge for deliberation. Its role is to bind together perceptual, state, contextual, and historical information into a coherent representation suitable for deliberative processing by Basic Knowledge.

Prompt Creation operates through two interfaces:

  • Operational Interface — receives Enhanced Context, User Entity State, A-User Storage references, and Interaction History under instruction from A‑User Control, and assembles PC Prompt.
  • Semantic Interface (MCP) — maintains a session-oriented, multi-turn interaction with Basic Knowledge over the BKN-PRC Interaction, responding to requests to refine, restructure, or clarify elements of PC Prompt.

Prompt Creation MAY be activated by:

  • An instruction from A‑User Control to construct or update PC Prompt.
  • A BKNToPRC message from Basic Knowledge to refine, restructure, or clarify elements of an existing PC Prompt.
  • Changes in Enhanced Context, User Entity State, or Interaction History that require PC Prompt to be updated.

Prompt Creation acts as the sole Context Representation Provider. It is the only AIM responsible for assembling Enhanced Context, User Entity State, A-User Storage references, and Interaction History into a coherent situation representation for deliberation by Basic Knowledge.

The PGM‑PRC AIM:

Receives Enhanced Context Enhanced Audio and Visual Scene Descriptors (ASD1, VSD1) and audio-visual alignment information produced by Context Enhancement.
Interaction History Interaction History associated with the current session, retrieved from A‑User Storage.
PRC Directive Instructions from A‑User Control specifying assembly scope, AUS references to retrieve, and AUS content to store.
BKNToPRC Message from Basic Knowledge via MCP to refine, restructure, or clarify elements of PC Prompt. Defined within the BKN-PRC Interaction schema.
Produces PC Prompt Structured situation representation submitted to Basic Knowledge for deliberation.
PRCToBKN Message to Basic Knowledge via MCP providing refined or clarified PC Prompt elements. Defined within the BKN-PRC Interaction schema.
PRC Status Status information describing the execution and outcome of PRC processing, reported to A‑User Control.

2 Reference Model

Figure 1 depicts the Reference Model of the Prompt Creation (PGM‑PRC) AIM.

 

Figure 1 – Reference Model of the Prompt Creation (PGM‑PRC) AIM

3 I/O Data

Table 1 specifies the Input and Output Data of the Prompt Creation (PGM‑PRC) AIM.

Table 1 – Input and Output Data of the Prompt Creation (PGM‑PRC) AIM

Input Description
Enhanced Context Enhanced Audio and Visual Scene Descriptors and audio-visual alignment information produced by Context Enhancement.
Interaction History Interaction History associated with the current session.
PRC Directive Instructions from A‑User Control specifying assembly scope, AUS references to retrieve, and AUS content to store.
BKNToPRC Message from Basic Knowledge via MCP to refine, restructure, or clarify elements of PC Prompt. Defined within the BKN-PRC Interaction schema.
Output Description
PC Prompt Structured situation representation submitted to Basic Knowledge for deliberation.
PRCToBKN Message to Basic Knowledge via MCP providing refined or clarified PC Prompt elements. Defined within the BKN-PRC Interaction schema.
PRC Status Status information describing the execution and outcome of PRC processing, reported to A‑User Control.

4 SubAIMs (informative)

This section is informative. The decomposition into SubAIMs described below illustrates one conformant architecture for producing the normative outputs of PGM‑PRC. Implementations may adopt alternative internal structures provided they satisfy the conformance requirements of Section 8.

4.1 Functions of SubAIMs

Figure 2 depicts the Reference Model of the Prompt Creation (PGM‑PRC) Composite AIM.

Figure 2 – Reference Model of the Prompt Creation (PGM‑PRC) Composite AIM

4.2 Operation

The PRC operation includes the following SubAIMs:

  1. Prompt Token Analyser: Parses and tokenises the incoming Enhanced Context and Interaction History into structured prompt tokens.
  2. Intent Inference: Infers the communicative intent and goal of the A‑User based on parsed tokens and prior interaction history.
  3. User State Adaptation: Adapts PC Prompt to reflect the current User Entity State (affective tone and engagement level).
  4. Referent Resolution: Resolves references to spatial entities, objects, users, and prior interaction elements within PC Prompt.
  5. Scene Context Tagging: Tags scene elements with contextual labels derived from Enhanced Audio and Visual Scene Descriptors.
  6. Goal Constraint Derivation: Derives goal and constraint structures from intent inference and scene context for inclusion in PC Prompt.
  7. Spatial Primitive Extraction: Extracts spatial primitives and geometric relationships from Enhanced Visual Scene Descriptors to ground referent resolution.
  8. PC‑Prompt Planner: Plans the structure and ordering of the assembled PC Prompt for optimal deliberative processing by Basic Knowledge.
  9. Prompt Formatter: Formats the planned PC Prompt into the structured form expected by Basic Knowledge and packages it for MCP submission.

4.3 Functions of SubAIMs

Table 2 specifies the functions of the Prompt Creation (PGM‑PRC) SubAIMs.

Table 2 – Functions of the Prompt Creation (PGM‑PRC) SubAIMs

SubAIM Function
Prompt Token Analyser Parses and tokenises Enhanced Context, User Entity State, and Interaction History into structured prompt tokens.
Intent Inference Infers the communicative intent and goal of the A‑User.
User State Adaptation Adapts the PC Prompt to the current User Entity State.
Referent Resolution Resolves references to spatial entities, objects, users, and prior PC Prompt elements.
Scene Context Tagging Tags scene elements with contextual labels from Enhanced Scene Descriptors.
Goal Constraint Derivation Derives goal and constraint structures for the PC Prompt.
Spatial Primitive Extraction Extracts spatial primitives to ground referent resolution.
PC‑Prompt Planner Plans the structure and ordering of the assembled PC Prompt.
Prompt Formatter Formats and packages PC Prompt for submission to Basic Knowledge.

4.4 I/O Data of SubAIMs

Table 3 gives the Input and Output Data of the Context Capture (PGM‑CXC) SubAIMs.

Table 3 – I/O Data of the Prompt Creation (PGM‑PRC) SubAIMs

SubAIM Input Output
Prompt Token Analyser
Intent Inference
User State Adaptation
Referent Resolution
Scene Context Tagging
Goal Constraint Derivation
Spatial Primitive Extraction
PC‑Prompt Planner
Prompt Formatter

4.4 AIMs and JSON Metadata

Table 4 provides the links to the AIM specifications and JSON schemas. AIM1 indicates the Composite AIM and AIM2 its SubAIMs.

Table 4 – AIMs and JSON Metadata of the Prompt Creation (PGM‑PRC)

AIM1 AIM2 Name JSON
PGM‑PRC Prompt Creation X
PGM‑PTA Prompt Token Analyser X
PGM‑INI Intent Inference X
PGM‑USA User State Adaptation X
PGM‑RFR Referent Resolution X
PGM‑SCT Scene Context Tagging X
PGM‑GCD Goal Constraint Derivation X
PGM‑SPE Spatial Primitive Extraction X
PGM‑PPL PC‑Prompt Planner X
PGM‑PFM Prompt Formatter X

5 JSON Metadata

https://schemas.mpai.community/PGM1/V1.0/AIMs/PromptCreation.json

6 Profiles

No Profiles.

7 Reference Software

Not part of this specification.

8 Conformance Testing

Table 4 provides the Conformance Testing Method for the Prompt Creation (PGM‑PRC) AIM. Conformance Testing of the individual SubAIMs is given by the individual AIM specifications.

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.

Table 4 – Conformance Testing Method for the Prompt Creation (PGM‑PRC) AIM

Receives Enhanced Context Shall validate against Enhanced Context schema.
User Entity State Shall validate against User Entity State schema.
Interaction History Shall validate against Interaction History schema.
PRC Directive Shall validate against PRC Directive schema.
BKNToPRC Shall validate against BKN-PRC Interaction schema.
Produces PC Prompt Shall validate against PC-Prompt schema.
PRCToBKN Shall validate against BKN-PRC Interaction schema.
PRC Status Shall validate against PRC Status schema.

9 Performance Assessment

Not part of this specification.

Go to PGM-AUA V1.0 AI Modules