generated: '2026-08-13'
method: derived
source: >-
  openapi/_original/unify-data-openapi.json,
  openapi/_original/unify-sequences-openapi.json,
  openapi/_original/unify-tasks-openapi.json,
  openapi/_original/unify-analytics-openapi.json
  (harvested live 2026-08-13),
  https://docs.unifygtm.com/developers/api/data/standard-objects
docs: https://docs.unifygtm.com/developers/api/data/standard-objects
entities:
  - name: object
    description: A data-platform object (table). Standard objects ship built in
      (company, person, user); custom objects can be created via the Data API.
  - name: attribute
    description: A field on an object, with a type (text, url, email address,
      reference, datetime, etc.) and optional uniqueness.
  - name: attribute_option
    description: An allowed option value for a select-style attribute.
  - name: record
    description: A row of an object, addressed by id or found/upserted via
      unique attribute values.
  - name: company
    description: Standard object representing an organization; deduplicated by
      normalized unique domain.
  - name: person
    description: Standard object representing an individual, unique by email.
  - name: user
    description: Standard object representing a Unify user (record owners).
  - name: sequence
    description: An outbound sequence with ordered steps (Sequences API).
  - name: sequence_step
    description: A step within a sequence.
  - name: sequence_enrollment
    description: A person's enrollment in a sequence, with pause/resume state.
  - name: sequence_enrollment_step
    description: One executed step of a sequence enrollment; exported through the
      Bulk API rather than read individually.
  - name: task
    description: >-
      A unit of outreach work - an email, phone call, LinkedIn touch, or manual
      action item - tied to a person and assigned to a user, carrying its own
      type, status, priority and due date. Added by the Tasks API
      (api.unifygtm.com/tasks/v1), new since the previous derivation.
  - name: analytics_event
    description: >-
      An identify, page, or track event collected from a website or product by the
      Intent client and posted to the Analytics API. Not individually addressable;
      exported in bulk through event query jobs.
  - name: query_job
    description: >-
      Asynchronous Bulk API job. Five families now exist - object records, events,
      sequence enrollments, sequence enrollment steps, and tasks - each with the
      same create / list / get / cancel / results lifecycle.
relationships:
  - from: object
    to: attribute
    kind: has_many
    via: object_name (path /objects/{object_name}/attributes)
  - from: attribute
    to: attribute_option
    kind: has_many
    via: attribute path /objects/{object_name}/attributes/{attribute_id}/options
  - from: object
    to: record
    kind: has_many
    via: path /objects/{object_name}/records
  - from: person
    to: company
    kind: belongs_to
    via: company (Reference attribute; settable by match on unique values such
      as domain)
  - from: company
    to: user
    kind: belongs_to
    via: record_owner (Reference attribute)
  - from: person
    to: user
    kind: belongs_to
    via: record_owner (Reference attribute)
  - from: sequence
    to: sequence_step
    kind: has_many
    via: path /sequences/{sequence_id}/steps
  - from: sequence
    to: sequence_enrollment
    kind: has_many
    via: sequence_id filter on list_sequence_enrollments
  - from: sequence_enrollment
    to: person
    kind: belongs_to
    via: person_id
  - from: sequence_enrollment
    to: sequence_enrollment_step
    kind: has_many
    via: enrollment-step query-job results (no synchronous list endpoint)
  - from: task
    to: person
    kind: belongs_to
    via: person_id
  - from: task
    to: user
    kind: belongs_to
    via: assigned_user_id (nullable)
  - from: query_job
    to: sequence_enrollment
    kind: has_many
    via: enrollment query-job results pages
  - from: query_job
    to: sequence_enrollment_step
    kind: has_many
    via: enrollment-step query-job results pages
  - from: query_job
    to: record
    kind: has_many
    via: object-record query-job results pages (/objects/{object_name}/query-jobs)
  - from: query_job
    to: task
    kind: has_many
    via: task query-job results pages (/tasks/v1/tasks/query-jobs)
  - from: query_job
    to: analytics_event
    kind: has_many
    via: event query-job results pages (/data/v1/events/query-jobs)
unique_keys:
  company: domain (normalized, deduplicating)
  person: email
  task: id (uuid); no natural key
enums:
  TaskType: [EMAIL, ACTION_ITEM, PHONE_CALL, LINKEDIN_SEND_MESSAGE, LINKEDIN_SEND_CONNECTION_REQUEST,
    LINKEDIN_VIEW_PROFILE, REPLY_TO_EMAIL]
  TaskStatus: [NOT_READY, READY, COMPLETED, SKIPPED, CANCELED]
  TaskPriority: [LOW, MEDIUM, HIGH, URGENT]
identifiers:
  scheme: bare RFC 4122 uuid
  prefixed: false
  note: >-
    Unify uses unprefixed uuids for every entity id (a `uuid` component schema is
    shared across all four documents). There is no Stripe-style type prefix, so an
    id carries no information about which entity it addresses - a caller holding a
    loose id cannot tell a task from an enrollment from a record without knowing
    where it came from.
coverage:
  entities: 14
  relationships: 17
  derived_from_operations: 68