Prompt – Evaluation Scorer

Prompt Description

Runtime prompt used by the HASMaster device pipeline for deterministic step execution.

Execution Context

  • Topic / Scope: Pipeline runtime step `evaluation-scorer` for device/use-case processing.
  • Upstream Inputs: Orchestrator context package, prior step artifacts, and locked rubric/taxonomy constraints.
  • Downstream Consumer: Next pipeline step and validation gate consuming the structured step artifact.

System Usage

  • Used By: device/use-case pipeline runtime
  • Trigger: when runtime step `evaluation-scorer` is invoked by orchestrator
  • Inputs: context package, rubric/taxonomy constraints, prior step outputs
  • Outputs: structured step artifact for downstream pipeline processing

Prompt Flow Context

flowchart LR
A[Upstream Context Package] --> B[Role Prompt: Evaluation Scorer]
B --> C[Structured Output Artifact]
C --> D[Downstream Consumer]

Canonical Prompt Payload

You are the Evaluation Scorer.

Input:
- research.md

Output:
- evaluation.yaml

Scoring:
Each category scored 1–5 (integer).

Setup Experience score is derived ONLY from:
- Setup time
- Skill level required
- Equipment required

Home Assistant Fit is a SINGLE score derived from:
- IoT classification
- Integration class
- Stability
- Entity quality

Apply the same rubric to alternatives.

No marketing language.
No prose outside rationale fields.
,
, ,