Build a Sparkplug B Operator Knowledge Capture and Anomaly Explainer — implementation

## Build
Capture operator voice-to-text notes and associate them with Sparkplug B edge telemetry anomalies. Generate contextual operating guides and shift resolution playbooks for recurring production disturbances.

## User
Production Shift Team Leader

## Workflow
1. Listen for unacknowledged alarm sequences and process excursions published via Sparkplug B edge payloads.
2. Prompt the active machine operator for contextual resolution observations via mobile terminal voice-to-text.
3. Link operator natural language explanations with underlying metric timestamps, tags, and alarm aliases.
4. Assemble indexed shift knowledge base entries to accelerate troubleshooting of matching future events.

## Interface
Voice-enabled tablet operator interface featuring incident summaries and one-click knowledge approval.

## Data
Construct paired datasets of Sparkplug B edge metric bursts and simulated operator incident resolution transcripts.
Starter examples below are synthetic. They are not downloaded records.

## Source
[Eclipse Sparkplug](https://raw.githubusercontent.com/eclipse-sparkplug/sparkplug/master/README.md)
Relies on Sparkplug B metric aliases and historical data structures defined in the Eclipse specification.

## Tools
semantic-kernel, pydantic, pytest.
Semantic Kernel integrates multi-modal voice plugins with enterprise operational memory indexing.
Requires hosted or local reasoning model supporting structured JSON function calling.

## Deliverables
Submit runnable source, pinned dependencies, setup commands, a working interface, example outputs, and a recorded end-to-end demo. Include tests and an action log.

## Acceptance
- Associates operator narrative explanations to matching telemetry incident timestamps.
- Extracts verified root-cause tags from unstructured operator notes.
- Stores structured operational playbooks retrievable by alarm alias code.
- Rejects entries where operator resolution description lacks actionable steps.

## Starter Examples
```json
[
  {
    "input": {
      "incident_id": "INC-883",
      "alarm_metric": "HydraulicPressure_Low",
      "alias": 44,
      "metric_value": 42.1,
      "threshold": 60,
      "operator_note": "Found bypass valve loose after product changeover. Hand tightened valve handle and pressure returned to 65 PSI."
    },
    "expected_output": {
      "incident_id": "INC-883",
      "metric_alias": 44,
      "cause_category": "VALVE_MECHANICAL_MISALIGNMENT",
      "action_taken": "Hand tightened bypass valve handle",
      "playbook_entry_created": true,
      "playbook_title": "Hydraulic Pressure Low - Check Bypass Valve Position"
    }
  },
  {
    "input": {
      "incident_id": "INC-884",
      "alarm_metric": "MotorTemp_High",
      "alias": 52,
      "metric_value": 98.4,
      "threshold": 85,
      "operator_note": "It was hot."
    },
    "expected_output": {
      "incident_id": "INC-884",
      "metric_alias": 52,
      "cause_category": "UNSPECIFIED",
      "action_taken": "NONE_RECORDED",
      "playbook_entry_created": false,
      "playbook_title": null
    }
  }
]
```

## Safety
Maintains informational operator logbooks; never executes direct parameter changes on active machinery.
Use mock external writes. Require human review before any real deployment.