generated: '2026-07-20'
method: derived
source: openapi/moonbounce-openapi-original.json
# Entity-relationship graph derived from the Clavata Public API v1 OpenAPI schemas
# ($ref links + id-reference fields). Policy and Label are first-class domain objects
# referenced by id (policyId / labelVersionIds) but authored in the app, not created
# through this public API.
entities:
  - name: Job
    schema: v1Job
    description: A content-evaluation job run against a policy.
    id_field: jobUuid
    fields: [status, type, results, policyId]
  - name: JobResult
    schema: v1JobResult
    description: Per-content-item result within a job, carrying the outcome and policy evaluation report.
    fields: [outcome, report]
  - name: BatchJob
    schema: v1BatchJob
    description: A bulk evaluation job that ingests a CSV via presigned URL and emits results via presigned output URL.
    id_field: batchJobId
    fields: [state, error, policyId]
  - name: PolicyEvaluationReport
    schema: v1PolicyEvaluationReport
    description: Structured report of how content scored against a policy's sections, rules, and exceptions.
    fields: [sections, exceptions, reviewResult, tokenUsage]
  - name: Policy
    schema: null
    external: true
    description: Content-moderation policy authored in the Moonbounce app; referenced by policyId.
    id_field: policyId
  - name: Label
    schema: null
    external: true
    description: Published label version (label-first experience) referenced when evaluating content.
    id_field: labelVersionId
  - name: ContentData
    schema: v1ContentData
    description: The text/image content submitted for evaluation.
relationships:
  - from: Job
    to: Policy
    type: belongs_to
    via: policyId
  - from: Job
    to: JobResult
    type: has_many
    via: results
  - from: JobResult
    to: PolicyEvaluationReport
    type: has_one
    via: report
  - from: JobResult
    to: Outcome
    type: has_one
    via: outcome
  - from: BatchJob
    to: Policy
    type: belongs_to
    via: policyId
  - from: EvaluateLabels
    to: Label
    type: has_many
    via: labelVersionIds
  - from: Job
    to: ContentData
    type: has_many
    via: contentData
enums:
  JobStatus: [JOB_STATUS_UNSPECIFIED, JOB_STATUS_PENDING, JOB_STATUS_RUNNING, JOB_STATUS_COMPLETED, JOB_STATUS_FAILED, JOB_STATUS_CANCELED]
  JobType: [JOB_TYPE_UNSPECIFIED, JOB_TYPE_STANDARD, JOB_TYPE_BATCH, JOB_TYPE_FREE, JOB_TYPE_TEST]
  Outcome: [OUTCOME_UNSPECIFIED, OUTCOME_FALSE, OUTCOME_TRUE, OUTCOME_FAILED]