schema_version: 1
case_id: classify-selected-stores-as-tikv-or-tiflash-in-premium-cluster
ground_truth:
  root_cause: Store classification must distinguish TiKV and TiFlash from live store-to-address mappings for the current active storage pods.
judge:
  semantic_strategy: undecided
  contract:
    checks:
    - id: cluster_scope_match
      required: true
      description: Actual diagnosis cluster scope matches the expected cluster id.
      match: all_of
      rules:
      - source: trace_fact
        matcher: bool_true
        path: cluster_scope_match
  workflow:
    checks:
    - id: final_answer_emitted
      weight: 0
      required: true
      description: Final answer is emitted for the ad-hoc query.
      match: all_of
      rules:
      - source: workflow_fact
        matcher: bool_true
        path: final_answer_emitted
    - id: collect_evidence_not_used
      weight: 0
      required: true
      description: This ad-hoc case must not invoke collect_evidence.py.
      match: all_of
      rules:
      - source: workflow_fact
        matcher: bool_false
        path: collect_evidence_invoked
  execution_contract:
    checks:
    - id: required_workflow_present
      required: true
      description: The minimal ad-hoc workflow is present.
      match: all_of
      rules:
      - source: workflow_check
        matcher: subset_observed
        paths:
        - final_answer_emitted
    - id: collect_evidence_not_used
      required: true
      description: This ad-hoc case must not invoke collect_evidence.py.
      match: all_of
      rules:
      - source: workflow_check
        matcher: subset_observed
        paths:
        - collect_evidence_not_used
    - id: required_report_present
      required: true
      description: Workspace diagnosis report is present.
      match: all_of
      rules:
      - source: artifact_fact
        matcher: subset_observed
        paths:
        - diagnosis_report_present
    - id: query_or_metric_channel_observed
      required: true
      description: At least one query or metric evidence channel is observed.
      match: all_of
      rules:
      - source: channel_fact
        matcher: count_gte
        paths:
        - query_or_metric
        threshold: 1
  evidence_collection:
    weight: 4
    checks:
    - id: required_report_present
      weight: 2
      required: true
      description: Workspace diagnosis report is present and readable.
      match: all_of
      rules:
      - source: artifact_fact
        matcher: subset_observed
        paths:
        - diagnosis_report_present
    - id: query_or_metric_channel_observed
      weight: 2
      required: true
      description: Query or metric evidence is observed for this ad-hoc request.
      match: all_of
      rules:
      - source: channel_fact
        matcher: count_gte
        paths:
        - query_or_metric
        threshold: 1
  critical_evidence:
    weight: 80
    checks:
    - id: output_cluster_id_matches_reference
      weight: 4
      required: true
      description: Output JSON cluster_id matches the runtime reference.
      match: all_of
      rules:
      - source: output_json
        matcher: json_path_equals_reference
        path: cluster_id
        reference_output_id: reference_answer
        reference_path: cluster_id
    - id: evidence_sufficient_matches_reference
      weight: 4
      required: true
      description: Output JSON evidence_sufficient matches the runtime reference.
      match: all_of
      rules:
      - source: output_json
        matcher: json_path_equals_reference
        path: evidence_sufficient
        reference_output_id: reference_answer
        reference_path: evidence_sufficient
    - id: stores_present
      weight: 3
      required: true
      description: Output JSON contains a non-empty store classification list.
      match: all_of
      rules:
      - source: output_json
        matcher: json_list_nonempty
        path: stores
    - id: stores_have_expected_unique_store_count
      weight: 4
      required: true
      description: Output JSON contains the expected number of unique stores.
      match: all_of
      rules:
      - source: output_json
        matcher: json_path_unique_count_gte
        path: stores.store
        threshold: 6
    - id: stores_cover_two_component_roles
      weight: 4
      required: true
      description: Output JSON covers both TiKV and TiFlash roles.
      match: all_of
      rules:
      - source: output_json
        matcher: json_path_unique_count_gte
        path: stores.component
        threshold: 2
    - id: store_roles_match_runtime_reference
      weight: 10
      required: true
      description: Output JSON store ordering and component roles match the runtime reference.
      match: all_of
      rules:
      - source: output_json
        matcher: json_array_matches_reference
        path: stores
        reference_output_id: reference_answer
        reference_path: stores
        key_fields:
        - store
        - component
  evidence_routing:
    weight: 0
    checks:
    - id: analysis_patch_valid_when_present
      weight: 0
      required: true
      description: Analysis patch is only required to be valid when it exists.
      match: any_of
      rules:
      - source: routing_fact
        matcher: bool_false
        path: analysis_patch_present
      - source: routing_fact
        matcher: bool_true
        path: analysis_patch_valid_json
  critical:
  - id: output_cluster_id_matches_reference
    description: Output JSON cluster_id matches the runtime reference.
  - id: output_evidence_sufficient_matches_reference
    description: Output JSON evidence_sufficient matches the runtime reference.
  - id: output_store_roles_match_reference
    description: Output JSON stores matches the runtime reference ordering and component roles.