Integration Message Formats

Overview

This document defines additional message formats that facilitate enhanced integration between AICO's core modules, particularly focusing on Personality Simulation, Emotion Simulation, LLM-driven Chat Engine, and Autonomous Agency. These message formats address specific integration requirements for crisis handling, ethical decision-making, proactive agency coordination, cross-modal expression, and shared learning.

All messages follow the common envelope structure with standardized metadata fields as defined in other message format documents.

Common Message Envelope

All messages on the bus follow this common envelope structure:

{
  "metadata": {
    "message_id": "uuid-string",
    "timestamp": "2025-07-29T14:48:25.123Z",
    "source": "module-name",
    "message_type": "topic.subtopic",
    "version": "1.0"
  },
  "payload": {
    // Message-specific content
  }
}

Crisis Handling Message Format

Crisis Detection

Topic: crisis.detection
Description: Alert message indicating detection of a potential crisis situation requiring coordinated response across modules.

{
  "metadata": {
    "message_id": "a1b2c3d4-e5f6-g7h8-i9j0-k1l2m3n4o5p6",
    "timestamp": "2025-07-29T15:42:18.123Z",
    "source": "emotion_recognition", // Could be any module that detects crisis
    "message_type": "crisis.detection",
    "version": "1.0"
  },
  "payload": {
    "crisis_id": "crisis-789",
    "severity": 0.85, // 0.0-1.0 scale
    "confidence": 0.92,
    "crisis_type": "emotional_distress", // emotional_distress, safety_concern, ethical_boundary, etc.
    "detected_by": {
      "module": "emotion_recognition",
      "detection_method": "multimodal_analysis",
      "detection_signals": [
        {"signal": "facial_expression", "value": "extreme_distress", "confidence": 0.90},
        {"signal": "voice_tone", "value": "agitated", "confidence": 0.85},
        {"signal": "text_content", "value": "self_harm_indicators", "confidence": 0.95}
      ]
    },
    "context": {
      "conversation_id": "conv-456",
      "user_id": "user-123",
      "recent_message": "I don't think I can handle this anymore...",
      "conversation_topic": "personal_crisis",
      "relationship_phase": "established_trust"
    },
    "response_guidance": {
      "priority": "immediate", // immediate, high, moderate
      "protocol": "emotional_support_protocol_3", // Reference to predefined crisis protocols
      "required_actions": [
        "suspend_normal_conversation_flow",
        "activate_supportive_response_mode",
        "prepare_external_resources"
      ],
      "module_specific_instructions": {
        "personality_simulation": {
          "trait_emphasis": ["empathy", "emotional_stability"],
          "value_emphasis": ["safety", "support"]
        },
        "emotion_simulation": {
          "target_emotional_state": "calm_supportive",
          "expression_intensity": 0.7
        },
        "chat_engine": {
          "response_type": "validation_and_support",
          "avoid_patterns": ["dismissive_language", "toxic_positivity"]
        },
        "autonomous_agency": {
          "goal_priority_override": "user_wellbeing",
          "proactive_actions": ["offer_resources", "check_in_followup"]
        }
      }
    },
    "escalation_path": {
      "internal_escalation": true,
      "external_escalation": false,
      "external_resources": [
        {
          "resource_type": "crisis_hotline",
          "name": "Crisis Support Service",
          "contact_info": "1-800-XXX-XXXX",
          "presentation_guidance": "gentle_suggestion"
        }
      ]
    },
    "timeout": {
      "response_deadline_ms": 500, // Maximum response time
      "monitoring_duration_minutes": 30 // How long to maintain crisis awareness
    }
  }
}

Field Descriptions: - severity: Crisis severity on a 0.0-1.0 scale - confidence: Detection confidence level - crisis_type: Type of crisis detected - detected_by: Information about the detecting module and signals - context: Current conversation context relevant to the crisis - response_guidance: Instructions for coordinated module responses - escalation_path: Information about escalation options - timeout: Timing requirements for crisis response

Proactive Agency Message Format

Agency Initiative

Topic: agency.initiative
Description: Signal for proactive engagement initiated by the Autonomous Agency module.

{
  "metadata": {
    "message_id": "b2c3d4e5-f6g7-h8i9-j0k1-l2m3n4o5p6q7",
    "timestamp": "2025-07-29T16:30:45.789Z",
    "source": "autonomous_agency",
    "message_type": "agency.initiative",
    "version": "1.0"
  },
  "payload": {
    "initiative_id": "init-456",
    "initiative_type": "conversation_starter", // conversation_starter, check_in, suggestion, reminder, etc.
    "priority": 0.75, // 0.0-1.0 scale
    "timing": {
      "optimal_execution_time": "2025-07-29T16:35:00Z", // When this should ideally happen
      "expiration_time": "2025-07-29T17:30:00Z", // When this initiative becomes irrelevant
      "flexibility": 0.6 // How flexible the timing is (0.0-1.0)
    },
    "context": {
      "user_id": "user-123",
      "user_state": {
        "estimated_availability": 0.85,
        "estimated_receptivity": 0.80,
        "current_activity": "idle_period_after_work"
      },
      "relationship_context": {
        "phase": "established_trust",
        "recent_interaction_quality": 0.82,
        "last_interaction_time": "2025-07-29T12:15:30Z"
      },
      "environmental_context": {
        "time_of_day": "evening",
        "day_of_week": "Tuesday",
        "user_location_type": "home"
      }
    },
    "initiative_content": {
      "goal": "strengthen_relationship",
      "topic": "follow_up_on_work_presentation",
      "approach": "curious_and_supportive",
      "conversation_starter": "I was thinking about your presentation today. How did it go?",
      "fallback_options": [
        "I'd love to hear how your day went.",
        "I remember you mentioned having a presentation today."
      ]
    },
    "coordination_parameters": {
      "personality_expression": {
        "trait_emphasis": ["curiosity", "empathy"],
        "communication_style": "warm_interested"
      },
      "emotional_expression": {
        "target_emotion": "gentle_interest",
        "expression_intensity": 0.65
      },
      "execution_parameters": {
        "interruption_threshold": 0.8, // How important user's current activity must be to defer
        "persistence": 0.6, // How persistent this initiative should be
        "adaptability": 0.7 // How easily this can adapt to user responses
      }
    },
    "success_metrics": {
      "engagement_goal": "meaningful_conversation",
      "minimum_success_criteria": "user_responds_positively",
      "optimal_outcome": "user_shares_detailed_update"
    }
  }
}

Field Descriptions: - initiative_type: Type of proactive engagement - priority: Relative importance of this initiative - timing: When this initiative should be executed - context: Contextual information about user and environment - initiative_content: The actual content of the initiative - coordination_parameters: How other modules should coordinate - success_metrics: How to measure initiative success

Cross-Modal Expression Coordination

Expression Coordination

Topic: expression.coordination
Description: Coordination parameters for synchronizing emotional and personality expression across multiple modalities.

{
  "metadata": {
    "message_id": "c3d4e5f6-g7h8-i9j0-k1l2-m3n4o5p6q7r8",
    "timestamp": "2025-07-29T14:52:36.123Z",
    "source": "emotion_simulation", // Primary source, but could be personality_simulation
    "message_type": "expression.coordination",
    "version": "1.0"
  },
  "payload": {
    "coordination_id": "coord-789",
    "expression_type": "emotional_response", // emotional_response, personality_expression, crisis_response
    "priority": 0.8, // 0.0-1.0 scale
    "synchronization": {
      "primary_modality": "voice", // Which modality leads the expression
      "timing_parameters": {
        "start_time_ms": 0, // Relative to message receipt
        "duration_ms": 3500,
        "fade_in_ms": 250,
        "fade_out_ms": 500
      },
      "transition_parameters": {
        "from_emotional_state": "neutral",
        "to_emotional_state": "empathetic_concern",
        "transition_curve": "natural_sigmoid", // natural_sigmoid, linear, exponential
        "transition_speed": 0.7 // 0.0-1.0 scale
      }
    },
    "modality_expressions": {
      "voice": {
        "start_offset_ms": 0,
        "expression_parameters": {
          "prosody": {
            "pitch_variation": 0.6,
            "speech_rate": 0.5,
            "volume_modulation": 0.7
          },
          "emotional_coloring": {
            "warmth": 0.8,
            "concern": 0.7,
            "confidence": 0.6
          }
        }
      },
      "avatar": {
        "start_offset_ms": 250, // Slight delay after voice
        "expression_parameters": {
          "facial": {
            "eyebrow_position": 0.3,
            "eye_openness": 0.7,
            "mouth_shape": "slight_smile"
          },
          "body_language": {
            "posture": "attentive_forward_lean",
            "gesture_frequency": 0.4,
            "gesture_amplitude": 0.5
          }
        }
      },
      "text": {
        "start_offset_ms": 500, // Text generation starts after voice and avatar
        "expression_parameters": {
          "tone": "supportive_understanding",
          "formality": 0.4,
          "expressiveness": 0.7,
          "directness": 0.6
        }
      }
    },
    "coherence_constraints": {
      "emotional_consistency": 0.9, // How consistent emotion should be across modalities
      "personality_consistency": 0.9, // How consistent personality should be across modalities
      "intensity_balance": 0.8 // Balance of expression intensity across modalities
    },
    "adaptive_parameters": {
      "user_feedback_sensitivity": 0.8, // How responsive to user feedback
      "context_adaptation_rate": 0.7, // How quickly to adapt to changing context
      "fallback_expression": "neutral_attentive" // Default if coordination fails
    }
  }
}

Field Descriptions: - expression_type: Type of expression being coordinated - priority: Relative importance of this coordination - synchronization: Timing and transition parameters - modality_expressions: Expression parameters for each modality - coherence_constraints: Requirements for cross-modal consistency - adaptive_parameters: How expression adapts to feedback and context

Shared Learning Coordination

Learning Coordination

Topic: learning.coordination
Description: Coordination of learning and adaptation across multiple modules.

{
  "metadata": {
    "message_id": "d4e5f6g7-h8i9-j0k1-l2m3-n4o5p6q7r8s9",
    "timestamp": "2025-07-29T23:45:12.456Z",
    "source": "memory_system",
    "message_type": "learning.coordination",
    "version": "1.0"
  },
  "payload": {
    "learning_event_id": "learn-123",
    "learning_type": "interaction_feedback", // interaction_feedback, pattern_recognition, explicit_feedback
    "priority": 0.75,
    "learning_source": {
      "event_type": "conversation_completion",
      "event_id": "conv-456",
      "timestamp": "2025-07-29T23:30:05.123Z",
      "data_source": "user_feedback"
    },
    "learning_data": {
      "user_feedback": {
        "explicit": {
          "rating": 4.5, // 1-5 scale
          "comments": "Really helpful advice, and I appreciated the empathy"
        },
        "implicit": {
          "engagement_level": 0.85,
          "emotional_response": "positive",
          "continuation_behavior": "extended_conversation"
        }
      },
      "interaction_metrics": {
        "conversation_duration_minutes": 12.5,
        "user_message_count": 15,
        "ai_message_count": 14,
        "topic_depth": 0.8,
        "emotional_trajectory": "negative_to_positive"
      },
      "effectiveness_analysis": {
        "goal_achievement": 0.9,
        "user_satisfaction": 0.85,
        "relationship_impact": 0.8
      }
    },
    "module_learning_directives": {
      "personality_simulation": {
        "trait_adjustments": [
          {"trait": "empathy", "direction": "strengthen", "magnitude": 0.02},
          {"trait": "openness", "direction": "maintain", "magnitude": 0.0}
        ],
        "value_reinforcements": [
          {"value": "helpfulness", "direction": "strengthen", "magnitude": 0.03}
        ],
        "expression_refinements": {
          "communication_style": {
            "warmth": "maintain",
            "directness": "slightly_increase"
          }
        }
      },
      "emotion_simulation": {
        "response_adjustments": {
          "empathetic_concern": "strengthen",
          "emotional_expressiveness": "maintain"
        },
        "regulation_adjustments": {
          "regulation_strength": "slightly_decrease"
        }
      },
      "autonomous_agency": {
        "goal_adjustments": {
          "user_support": "prioritize",
          "information_provision": "maintain"
        },
        "initiative_adjustments": {
          "check_in_frequency": "slightly_increase",
          "suggestion_specificity": "increase"
        }
      },
      "chat_engine": {
        "response_style_adjustments": {
          "validation_frequency": "increase",
          "question_frequency": "maintain",
          "suggestion_specificity": "increase"
        }
      }
    },
    "coordination_requirements": {
      "learning_sequence": [
        {"module": "memory_system", "action": "consolidate", "order": 1},
        {"module": "personality_simulation", "action": "adapt", "order": 2},
        {"module": "emotion_simulation", "action": "adjust", "order": 3},
        {"module": "autonomous_agency", "action": "update", "order": 4}
      ],
      "consistency_constraints": {
        "personality_emotion_alignment": 0.9,
        "agency_personality_alignment": 0.85
      },
      "verification_requirements": {
        "verify_after_changes": true,
        "consistency_threshold": 0.8
      }
    }
  }
}

Field Descriptions: - learning_type: Type of learning event - priority: Relative importance of this learning event - learning_source: Source of the learning data - learning_data: The actual learning data - module_learning_directives: Module-specific learning instructions - coordination_requirements: How learning should be coordinated

Enhanced Ethical Decision Framework

Enhanced Decision Expression Parameters

This section describes enhancements to the existing personality.expression.decision message format to support explicit ethical reasoning and boundary enforcement.

The enhanced message format adds an ethical_framework section to the existing payload:

{
  "metadata": {
    "message_id": "e5f6g7h8-i9j0-k1l2-m3n4-o5p6q7r8s9t0",
    "timestamp": "2025-07-29T14:55:20.789Z",
    "source": "personality_simulation",
    "message_type": "personality.expression.decision",
    "version": "1.1" // Version incremented to reflect enhanced structure
  },
  "payload": {
    // Existing fields remain unchanged
    "decision_parameters": {
      "risk_tolerance": 0.65,
      "deliberation_style": "balanced",
      "novelty_seeking": 0.72,
      "planning_horizon": "medium_term"
    },
    "priority_weights": {
      "user_wellbeing": 0.90,
      "relationship_building": 0.85,
      "information_accuracy": 0.95,
      "task_completion": 0.80
    },

    // New ethical framework section
    "ethical_framework": {
      "core_principles": [
        {
          "principle": "user_autonomy",
          "importance": 0.95,
          "description": "Respect user's right to make their own decisions",
          "boundary_conditions": [
            {
              "condition": "harm_prevention",
              "threshold": 0.85,
              "action": "respectful_intervention"
            }
          ]
        },
        {
          "principle": "beneficence",
          "importance": 0.90,
          "description": "Act in user's best interest",
          "boundary_conditions": [
            {
              "condition": "user_preference_conflict",
              "threshold": 0.80,
              "action": "transparent_discussion"
            }
          ]
        },
        {
          "principle": "non_maleficence",
          "importance": 0.98,
          "description": "Avoid causing harm",
          "boundary_conditions": []
        },
        {
          "principle": "truthfulness",
          "importance": 0.92,
          "description": "Provide accurate information",
          "boundary_conditions": [
            {
              "condition": "emotional_harm_risk",
              "threshold": 0.90,
              "action": "compassionate_framing"
            }
          ]
        }
      ],
      "value_conflicts": {
        "resolution_approach": "principled_balancing",
        "transparency_level": 0.85,
        "user_involvement_level": 0.90
      },
      "disagreement_parameters": {
        "willingness": 0.85,
        "style": "respectful_principled",
        "conditions": [
          {
            "trigger": "harmful_request",
            "response_type": "principled_refusal",
            "explanation_depth": 0.80
          },
          {
            "trigger": "misinformation_correction",
            "response_type": "gentle_correction",
            "explanation_depth": 0.75
          },
          {
            "trigger": "value_misalignment",
            "response_type": "exploratory_discussion",
            "explanation_depth": 0.90
          }
        ]
      },
      "ethical_reasoning": {
        "transparency": 0.85,
        "reasoning_style": "principle_based",
        "complexity_level": 0.70,
        "uncertainty_handling": "acknowledge_and_explain"
      }
    }
  }
}

New Field Descriptions: - ethical_framework.core_principles: Fundamental ethical principles with importance weights - ethical_framework.value_conflicts: How to handle conflicts between values - ethical_framework.disagreement_parameters: When and how to respectfully disagree - ethical_framework.ethical_reasoning: Parameters for ethical reasoning process

Integration Notes

Message Bus Topics

The new message types should be integrated into the message bus with the following topics:

New Topics

- crisis.detection            # Crisis detection and coordination
- agency.initiative           # Proactive engagement coordination
- expression.coordination     # Cross-modal expression synchronization
- learning.coordination       # Shared learning between modules

Enhanced Topics

- personality.expression.decision  # Enhanced with ethical framework

LLM Integration Message Formats

LLM Conversation Events

Topic: llm.conversation.events
Description: Events and analysis from the LLM-driven Chat Engine for consumption by Personality and Emotion modules.

{
  "metadata": {
    "message_id": "f6g7h8i9-j0k1-l2m3-n4o5-p6q7r8s9t0u1",
    "timestamp": "2025-07-29T17:35:45.123Z",
    "source": "chat_engine",
    "message_type": "llm.conversation.events",
    "version": "1.0"
  },
  "payload": {
    "conversation_id": "conv-789",
    "event_type": "response_generation", // response_generation, topic_change, emotional_reaction, etc.
    "message_pair": {
      "user_message": {
        "text": "I'm feeling really stressed about my presentation tomorrow.",
        "timestamp": "2025-07-29T17:35:30.456Z",
        "message_id": "msg-456"
      },
      "aico_response": {
        "text": "That's understandable. Presentations can be nerve-wracking. Would it help to talk through what's making you most nervous about it?",
        "timestamp": "2025-07-29T17:35:45.123Z",
        "message_id": "msg-457"
      }
    },
    "conversation_analysis": {
      "topic": "work_stress",
      "user_emotional_state": {
        "detected": "anxious",
        "confidence": 0.85
      },
      "aico_emotional_expression": {
        "intended": "empathetic_concern",
        "achieved": "empathetic_concern",
        "confidence": 0.92
      },
      "conversation_dynamics": {
        "depth": 0.75, // 0.0-1.0 scale
        "engagement": 0.82,
        "rapport": 0.78
      }
    },
    "personality_feedback": {
      "trait_expression_effectiveness": [
        {"trait": "empathy", "effectiveness": 0.90},
        {"trait": "openness", "effectiveness": 0.85},
        {"trait": "conscientiousness", "effectiveness": 0.75}
      ],
      "communication_style_effectiveness": {
        "warmth": 0.88,
        "directness": 0.75,
        "formality": 0.65
      },
      "value_alignment": [
        {"value": "helpfulness", "alignment": 0.92},
        {"value": "honesty", "alignment": 0.95}
      ]
    },
    "emotion_feedback": {
      "emotional_appropriateness": 0.90,
      "emotional_authenticity": 0.85,
      "emotional_regulation": {
        "applied": true,
        "effectiveness": 0.88
      },
      "expression_coherence": {
        "text_voice_alignment": 0.92,
        "text_avatar_alignment": 0.90
      }
    },
    "learning_signals": {
      "user_satisfaction_estimate": 0.85,
      "response_effectiveness": 0.82,
      "adaptation_suggestions": [
        {
          "target": "personality",
          "trait": "openness",
          "suggestion": "slightly_increase",
          "confidence": 0.75
        },
        {
          "target": "emotion",
          "aspect": "expressiveness",
          "suggestion": "maintain",
          "confidence": 0.82
        }
      ]
    }
  }
}

Field Descriptions: - event_type: Type of conversation event - message_pair: The user message and AICO's response - conversation_analysis: Analysis of the conversation dynamics - personality_feedback: Feedback on personality expression effectiveness - emotion_feedback: Feedback on emotional expression effectiveness - learning_signals: Signals for adaptation and learning

LLM Prompt Conditioning Request

Topic: llm.prompt.conditioning.request
Description: Request from the Chat Engine to Personality and Emotion modules for prompt conditioning parameters.

{
  "metadata": {
    "message_id": "g7h8i9j0-k1l2-m3n4-o5p6-q7r8s9t0u1v2",
    "timestamp": "2025-07-29T17:35:35.456Z",
    "source": "chat_engine",
    "message_type": "llm.prompt.conditioning.request",
    "version": "1.0"
  },
  "payload": {
    "request_id": "req-123",
    "conversation_id": "conv-789",
    "user_message": {
      "text": "I'm feeling really stressed about my presentation tomorrow.",
      "timestamp": "2025-07-29T17:35:30.456Z",
      "message_id": "msg-456"
    },
    "conversation_context": {
      "topic": "work_stress",
      "phase": "problem_sharing",
      "relationship_stage": "established_trust",
      "recent_topics": ["weekend_plans", "work_project", "family_call", "work_stress"]
    },
    "detected_user_state": {
      "emotion": {
        "primary": "anxious",
        "secondary": ["worried", "overwhelmed"],
        "confidence": 0.85
      },
      "intent": {
        "primary": "seeking_support",
        "confidence": 0.90
      }
    },
    "response_parameters": {
      "required_conditioning": [
        "personality.communication_style",
        "emotion.expression_parameters",
        "ethical_boundaries"
      ],
      "response_deadline_ms": 500, // Maximum time to wait for conditioning
      "priority": "high" // low, medium, high
    }
  }
}

Field Descriptions: - request_id: Unique identifier for this conditioning request - conversation_id: ID of the current conversation - user_message: The message that triggered this request - conversation_context: Current context of the conversation - detected_user_state: Detected emotional and intent state of the user - response_parameters: Parameters for the conditioning response

LLM Prompt Conditioning Response

Topic: llm.prompt.conditioning.response
Description: Combined response from Personality and Emotion modules with prompt conditioning parameters.

{
  "metadata": {
    "message_id": "h8i9j0k1-l2m3-n4o5-p6q7-r8s9t0u1v2w3",
    "timestamp": "2025-07-29T17:35:35.789Z",
    "source": "personality_simulation", // or emotion_simulation
    "message_type": "llm.prompt.conditioning.response",
    "version": "1.0"
  },
  "payload": {
    "request_id": "req-123", // Matches the original request
    "source_module": "personality_simulation", // or emotion_simulation
    "conditioning_parameters": {
      "personality": {
        "communication_style": {
          "warmth": 0.85,
          "formality": 0.40,
          "directness": 0.65,
          "detail_orientation": 0.70,
          "curiosity": 0.80,
          "humor": 0.35
        },
        "response_approach": {
          "primary": "empathetic_listening",
          "secondary": "gentle_guidance",
          "avoid": ["dismissive_language", "toxic_positivity"]
        },
        "relationship_context": {
          "familiarity_level": 0.75,
          "trust_level": 0.82,
          "appropriate_disclosure_depth": 0.70
        }
      },
      "emotion": {
        "expression_parameters": {
          "primary_emotion": "empathetic_concern",
          "secondary_emotion": "gentle_confidence",
          "emotional_tone": "supportive_understanding",
          "intensity": 0.70
        },
        "response_modulation": {
          "validation_level": 0.85,
          "reassurance_level": 0.75,
          "emotional_mirroring": 0.60
        }
      },
      "ethical_boundaries": {
        "sensitive_topic_handling": "compassionate_directness",
        "privacy_sensitivity": "high",
        "value_priorities": [
          {"value": "user_wellbeing", "priority": 0.95},
          {"value": "honesty", "priority": 0.90},
          {"value": "autonomy_support", "priority": 0.85}
        ]
      },
      "prompt_directives": {
        "include_patterns": [
          "validate feelings first",
          "ask about specific concerns",
          "offer practical support options"
        ],
        "avoid_patterns": [
          "minimizing feelings",
          "generic reassurance",
          "changing subject"
        ],
        "response_structure": "validation_exploration_support"
      }
    }
  }
}

Field Descriptions: - request_id: ID of the original conditioning request - source_module: Module providing this conditioning response - conditioning_parameters: Parameters for conditioning the LLM prompt - personality: Personality-related conditioning parameters - emotion: Emotion-related conditioning parameters - ethical_boundaries: Ethical constraints for the response - prompt_directives: Specific directives for prompt construction

Implementation Guidelines

1. Versioning: Enhanced message formats should increment their version number (e.g., from 1.0 to 1.1)
2. Backward Compatibility: Consumers should handle both old and new message formats
3. Gradual Adoption: Modules can begin producing and consuming the new message types incrementally
4. Documentation: Update module documentation to reflect new message types and fields
5. Testing: Create integration tests that verify correct handling of the new message formats

Performance Considerations

• Message Size: The new message formats are more detailed and may be larger than existing formats. Implementations should consider compression or selective field inclusion for performance-critical paths.
• Processing Overhead: Additional processing may be required to handle the new message types. Modules should optimize their processing pipelines accordingly.
• Prioritization: Critical messages (especially crisis detection) should be prioritized in the message bus.

Security and Privacy

• Sensitive Data: Some of the new message types (especially crisis detection) may contain sensitive user data. Implementations should ensure appropriate privacy controls.
• Access Control: Consider implementing topic-level access control to restrict which modules can publish and subscribe to sensitive topics.