generated: '2026-08-12'
method: derived
source: openapi/_original/ahrefs-openapi-original.json
note: >-
  Ahrefs' API is a reporting surface, not a CRUD resource graph. 144 component schemas exist but 118 of
  the 148 operations are report readers whose "schema" is a row shape for one report, not a persistent
  entity — there is no `GET /backlinks/{id}`, no self link, and no id on a backlink row. Only the
  Management, Site Audit and Social Media families expose durable, id-addressed objects. The graph
  below therefore separates PROJECT-SCOPED ENTITIES (real objects with ids and lifecycles) from
  ANALYTICAL SUBJECTS (targets, keywords, brands — addressed by value, never by id) and REPORT ROW
  SHAPES (per-report projections).
entities:
- name: project
  id_field: project_id
  id_referenced_by_operations: 63
  domain: Management / Rank Tracker / Site Audit / Web Analytics / GSC Insights / Bot Analytics
  lifecycle: create, read, update, delete
  operations:
  - management.projects
  - management.post-projects
  - management.delete-projects
  fields_seen:
  - project_id
  - verified
  - folder
  note: >-
    The single most-referenced identifier in the API and the spine of everything account-scoped. A
    project can be placed in a folder (`folder_id` on create, `folder` returned on read) and carries a
    `verified` flag.
- name: project_keyword
  id_field: null
  parent: project
  domain: Rank Tracker
  operations:
  - management.project-keywords
  - management.project-keywords-delete
  - management.project-keywords-tags
  - management.project-keywords-tags-delete
  note: Keywords tracked inside a project; addressed by project_id + keyword, and tagged via PUT operations.
- name: project_competitor
  id_field: null
  parent: project
  domain: Rank Tracker
  operations:
  - management.project-competitors
  - management.project-competitors-delete
- name: keyword_list
  id_field: keyword_list_id
  id_referenced_by_operations: 8
  domain: Keywords Explorer / Management / GSC Insights
  operations:
  - management.keyword-list-keywords
  - management.keyword-list-keywords-delete
  note: >-
    Referencing a keyword_list_id in a Keywords Explorer request always makes that request billable,
    even when the keywords themselves are free test values.
- name: brand_radar_report
  id_field: report_id
  id_referenced_by_operations: 12
  domain: Brand Radar
  lifecycle: create, read, update
  operations:
  - management.brand-radar-reports
  - management.post-brand-radar-reports
  - management.patch-brand-radar-reports
  note: Saved Brand Radar report configuration, including brands, competitors and URL groups.
- name: brand_radar_prompt
  id_field: null
  parent: brand_radar_report
  domain: Brand Radar
  operations:
  - management.brand-radar-prompts
  - management.brand-radar-prompts-delete
- name: portfolio
  id_field: portfolio_id
  id_referenced_by_operations: 11
  domain: Site Explorer
  note: A saved collection of targets used as an alternative to `target` on Site Explorer reports.
- name: location
  id_field: null
  domain: Management
  operations:
  - management.locations
  note: Reference data — the geo/language locations a Rank Tracker project can track.
- name: social_channel
  id_field: channel
  domain: Social Media
  operations:
  - social-media.channels
  - social-media.channel-metrics
- name: social_post
  id_field: post
  parent: social_channel
  domain: Social Media
  lifecycle: create, read, update, delete
  operations:
  - social-media.posts
  - social-media.post
  - social-media.post-metrics
  - social-media.patch-post
  - social-media.delete-post
  note: The only fully CRUD-able content object in the API.
- name: audited_page
  id_field: null
  parent: project
  domain: Site Audit
  operations:
  - site-audit.page-explorer
  - site-audit.page-content
  - site-audit.issues
  note: >-
    Addressed by project + URL. page-content returns page text, now also as Markdown (`page_text_md`);
    page-explorer returns crawl and AI-bot allow/block fields.
analytical_subjects:
- name: target
  addressed_by: value
  parameter: target
  used_by_operations: 30
  modifiers:
  - mode (exact | prefix | domain | subdomains)
  - protocol
  - date
  - date_compared
  domain: Site Explorer, Batch Analysis
  note: A domain or URL. No id, no persistent record — the unit of analysis, not an entity.
- name: keyword
  addressed_by: value
  parameters:
  - keyword
  - keywords
  modifiers:
  - country
  - search_volume_type
  - volume_mode
  domain: Keywords Explorer, SERP Overview, Rank Tracker, GSC Insights
- name: brand
  addressed_by: value
  parameters:
  - brand
  - prompts
  - data_source
  - market
  domain: Brand Radar
  note: >-
    Brand Radar's data_source enumerates AI engines (Claude and Grok were added in 2026, alongside
    google_ai_mode_keywords and google_ai_overviews_keywords).
relationships:
- from: project
  to: project_keyword
  type: has_many
  via: project_id
- from: project
  to: project_competitor
  type: has_many
  via: project_id
- from: project
  to: audited_page
  type: has_many
  via: project_id
- from: project_keyword
  to: project
  type: belongs_to
  via: project_id
- from: brand_radar_report
  to: brand_radar_prompt
  type: has_many
  via: report_id
- from: brand_radar_prompt
  to: brand_radar_report
  type: belongs_to
  via: report_id
- from: keyword_list
  to: keyword
  type: has_many
  via: keyword_list_id
- from: portfolio
  to: target
  type: has_many
  via: portfolio_id
- from: social_channel
  to: social_post
  type: has_many
  via: channel
- from: social_post
  to: social_channel
  type: belongs_to
  via: channel
- from: subscription
  to: project
  type: has_many
  via: workspace
  note: >-
    Implicit. subscription-info.limits-and-usage reports the workspace's API-unit allowance and usage;
    everything else is scoped inside that workspace.
report_row_shapes:
  count: 131
  note: >-
    Per-report projections such as site-explorer.all-backlinks, web-analytics.countries-chart and
    gsc.ctr-by-position. They share no common base schema, carry no id, and are not $ref'd by one
    another — components.schemas reuse across tools is effectively zero, which is why the spec is 2 MB.
  by_tool:
    site-explorer: 28
    web-analytics: 34
    brand-radar: 12
    gsc: 12
    management: 14
    social-media: 7
    keywords-explorer: 6
    rank-tracker: 6
    site-audit: 4
    public: 4
    batch-analysis: 2
    serp-overview: 1
    subscription-info: 1
id_conventions:
  prefixes: none
  format: 'no documented id format; ids are opaque and per-family (project_id, report_id, keyword_list_id,
    portfolio_id)'
  note: Ahrefs publishes no object-reference page documenting id shapes or prefixes.