> ## Documentation Index
> Fetch the complete documentation index at: https://docs.omnia-voice.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Get the persistent alignment report

> Reconstructed from the stored confusion rows: metrics with CIs, population breakdown, and every judge/human disagreement with the human critique and response excerpt. Free — reads the last run; POST /align re-judges and spends.



## OpenAPI

````yaml /openapi.json get /v1/criteria/{id}/alignment
openapi: 3.1.0
info:
  title: Omnia Management API
  description: >-
    The management API behind the improvement loop: capture and setup, request
    logs and datasets, grades (labels), judges (criteria), evals and deploy
    gates, fine-tuning and reinforcement learning, dedicated GPU endpoints, and
    model aliases and versions. Authenticated with a workspace API key
    (sk_sovereign_...). The inference API (chat, embeddings, rerank, responses)
    is OpenAI-compatible and documented separately.


    Responses are snake_case, list endpoints on the loop products use the
    {"object": "list", "data": [...]} envelope, and refusals use the same nested
    error shape the gateway emits: {"error": {"message", "type", "code"}}.
    Request bodies on the loop products (logs, labels, criteria, evals,
    datasets, aliases) are snake_case; the training and infrastructure products
    (fine-tuning, GRPO, environment tools, dedicated, model-version adoption)
    validate camelCase bodies, and each schema below says which it is. Endpoints
    that spend money require a key minted by a workspace owner or admin and
    return 403 otherwise.
  version: 1.0.0
servers:
  - url: https://gateway.omnia-voice.com
    description: Production
  - url: https://platform.omnia-voice.com/api
    description: Production (legacy alias — same API, older base URL)
security:
  - bearerAuth: []
paths:
  /v1/criteria/{id}/alignment:
    get:
      summary: Get the persistent alignment report
      description: >-
        Reconstructed from the stored confusion rows: metrics with CIs,
        population breakdown, and every judge/human disagreement with the human
        critique and response excerpt. Free — reads the last run; POST /align
        re-judges and spends.
      operationId: getCriteriaAlignment
      parameters:
        - name: id
          in: path
          required: true
          schema:
            type: string
      responses:
        '200':
          description: The persistent report, reconstructed from the stored confusion rows.
          content:
            application/json:
              schema:
                type: object
                properties:
                  aligned_at:
                    type:
                      - string
                      - 'null'
                  scope_tag:
                    type: string
                    description: The population measured ("" = all labeled traffic).
                  tag_breakdown:
                    type: array
                    description: What the measured labels were made of, by tag.
                    items:
                      type: object
                      properties:
                        tag:
                          type: string
                        'n':
                          type: integer
                  mixed_population:
                    type: boolean
                    description: >-
                      True when an unscoped criterion was measured across more
                      than one tag: the numbers describe a blend of tasks.
                  tier:
                    type: string
                    enum:
                      - aligned
                      - weak
                      - misaligned
                      - unmeasured
                  thin_alignment_set:
                    type: boolean
                    description: True under 100 judgeable labels; intervals stay wide.
                  metrics:
                    type: object
                    description: >-
                      n, tpr, tnr, kappa plus the 95% Wilson intervals. Note:
                      the interval keys on this payload are camelCase (tprCi,
                      tnrCi).
                  agreements:
                    type: integer
                  disagreements:
                    type: array
                    items:
                      type: object
                      properties:
                        request_id:
                          type: string
                        judge_verdict:
                          type: string
                        human_verdict:
                          type: string
                        critique:
                          type:
                            - string
                            - 'null'
                          description: The human's WHY from the label, when they wrote one.
                        tag:
                          type: string
                        response_excerpt:
                          type: string
                          description: >-
                            First ~300 chars of the judged response; "" when
                            content aged out.
        '404':
          description: No alignment run yet
      security:
        - bearerAuth: []
components:
  securitySchemes:
    bearerAuth:
      type: http
      scheme: bearer
      description: >-
        Your workspace API key, e.g. `sk_sovereign_...`, sent as `Authorization:
        Bearer <key>`.

````