# Vendor facets — DeveloperHub. A hosted docs and developer-portal platform: an OpenAPI reference
# with try-it, guides, changelogs, versioning, landing pages and AI tooling, server-side rendered.
# The headline: solid portal pointers and an auto-generated llms.txt. Its MCP server is for
# AUTHORS editing docs from Claude, Cursor or Codex, not for agents calling the API, so it earns
# nothing on mcp_server.
vendor: developerhub
name: DeveloperHub
website: https://developerhub.io
areas:
- developer-portal
registry_keys:
- developerhub
rubric_schema_version: 0.22.0
generated: '2026-09-25'
features_refreshed: '2026-09-25'
basis: capability
summary: >-
  DeveloperHub lifts the portal pointers: portal, documentation, API reference, changelog and a try-it
  console, each once declared in apis.yml. It renders server-side, so pages are readable without JavaScript,
  which avoids the SPA-shell trap that hides other portals. Its llms.txt is generated automatically, refreshes
  every six hours and cannot be edited (llms-full.txt is not supported). It counts only when served on
  the provider's own domain. Its MCP server is an authoring tool, so a provider on DeveloperHub has no
  reader-facing MCP server from it. Nothing here writes to the contract: the reference renders whatever
  examples and errors the OpenAPI has.
features:
- id: api-reference
  name: API reference from OpenAPI with try-it
  description: A reference generated from OpenAPI with a request playground supporting auth and OAuth
    2.0.
  source: https://developerhub.io/
  tier: all
- id: guides
  name: Guides and product documentation
  description: >-
    Authored guides with versioning, localization, a block-based landing page designer and GitHub two-way
    sync.
  source: https://developerhub.io/
  tier: all
- id: changelogs
  name: Changelogs
  description: Changelog pages as part of the docs project.
  source: https://developerhub.io/
  tier: all
- id: ssr-hosting
  name: Server-side rendering on a custom domain or subdirectory
  description: >-
    Docs are server-rendered and can be hosted on a custom domain or under a subdirectory behind a reverse
    proxy.
  source: https://docs.developerhub.io/support-center/hosting
  tier: all
- id: llms-txt
  name: Auto-generated llms.txt and Markdown pages
  description: >-
    Generates /llms.txt at the project base path every six hours (no llms-full.txt), and serves page Markdown
    at a .md suffix when enabled.
  source: https://docs.developerhub.io/support-center/llms-txt
  tier: all
- id: authoring-mcp
  name: MCP server for authoring
  description: >-
    Connects Claude, Cursor or Codex to the docs project so writers can read and write documentation from
    those tools.
  source: https://developerhub.io/ai
  tier: unknown
- id: ai-assistant
  name: AI Assistant for readers
  description: Reader-facing answers cited from the project's own pages.
  source: https://developerhub.io/ai
  tier: unknown
maps:
- feature: guides
  check: portal_present
  layer: composite
  provider_must: Declare the site as a DeveloperPortal entry in apis.yml.
  catalog_pass_rate: 0.228
  facet: developer_ergonomics
  points: 4
  baseline_pass_rate: 0.633
- feature: guides
  check: documentation_present
  layer: composite
  provider_must: Declare the guides as a Documentation entry.
  catalog_pass_rate: 0.453
  facet: developer_ergonomics
  points: 4
  baseline_pass_rate: 0.95
  saturated: true
  saturated_note: >-
    95% of providers with a contract, docs and a reference already earn this; the vendor cannot move it
    for most of its buyers.
- feature: api-reference
  check: api_reference_present
  layer: composite
  provider_must: Declare the reference as an APIReference entry.
  catalog_pass_rate: 0.222
  facet: developer_ergonomics
  points: 3
  baseline_pass_rate: 0.942
  saturated: true
  saturated_note: >-
    94% of providers with a contract, docs and a reference already earn this; the vendor cannot move it
    for most of its buyers.
- feature: api-reference
  check: console_or_sandbox
  layer: composite
  provider_must: Declare the try-it playground as a Console or Playground entry.
  catalog_pass_rate: 0.089
  facet: developer_ergonomics
  points: 3
  baseline_pass_rate: 0.332
- feature: changelogs
  check: change_log_present
  layer: composite
  provider_must: Maintain the changelog and declare it as a ChangeLog entry.
  catalog_pass_rate: 0.157
  facet: operational_transparency
  points: 6
  baseline_pass_rate: 0.438
- feature: llms-txt
  check: llms_txt_published
  layer: composite
  conditional: true
  condition: >-
    Only when the project is served on the provider's own domain (a custom domain or a subdirectory).
    The file sits at the project URL, and on a DeveloperHub-hosted URL it is not the provider's surface.
  note: >-
    It is generated from the provider's own docs and cannot be edited. On the provider's domain, format
    rule 2 credits it in full.
  catalog_pass_rate: 0.351
  facet: discoverability
  points: 4
  baseline_pass_rate: 0.65
earns_nothing:
- feature: authoring-mcp
  check: mcp_server
  layer: agent_readiness
  why: >-
    The server lets writers edit docs from an AI client. It does not expose the provider's API to agents.
- feature: ai-assistant
  check: agent_skills
  layer: agent_readiness
  why: A reader chat assistant is not a skills artifact.
- feature: api-reference
  check: examples_present
  why: The reference renders the examples the OpenAPI already has and adds none.
out_of_reach:
  checks:
  - sdk_count_1
  - sdk_count_3
  - cli_present
  - contract_present
  - overlay_published
  - status_page_present
  - rate_limits_documented
  - well_known_published
  - spec_presence
  note: >-
    A docs platform renders the contract. It does not generate SDKs or CLIs, or write, host or improve
    the OpenAPI.
unscored_practice:
- feature: llms-txt
  why: >-
    Per-page Markdown twins (.md) served to AI assistants while crawlers get HTML. The rubric has no markdown-twin
    check.
surface:
  discoverability:
    reachable: 4.0
    total: 54
  operational_transparency:
    reachable: 6.0
    total: 38
  developer_ergonomics:
    reachable: 14.0
    total: 42
  agent_readiness:
    reachable: 0
    total: 139
hard_rule: >-
  A model, not a score. Adopting this vendor changes a provider's Kin Score only when the provider publishes
  the resulting artifacts on its own surface; nothing here writes a score, and no sponsorship or partnership
  can.
method: searched
source:
- https://developerhub.io/
- https://developerhub.io/ai
- https://docs.developerhub.io/support-center/hosting
- https://docs.developerhub.io/support-center/llms-txt
measured:
  cohort:
    method: vendors-catalog.json detections (CNAME / header / URL shape / markup), never a name match
    detected: 1
    in_baseline: 0
  control:
    basis: providers earning contract_present + documentation_present + api_reference_present, minus the
      cohort
    n: 5216
  metric: >-
    cohort_pct / control_pct = mean share of the check's points earned (derived and platform credit weighted),
    x100
  measured_on: '2026-09-25'
  status: 'not measurable: 0 detected customers clear the baseline (need 20)'
simulation:
  simulated_on: '2026-09-25'
  rubric: 0.23.0
  population: providers publishing a contract (contract_present earned), replayable exactly
  providers: 8977
  providers_unreplayable: 987
  providers_moved: 8292
  conditional_rows: excluded (they depend on what the API already does)
  composite_lift:
    median: 4.7
    p75: 6.1
    p90: 6.8
    max: 8.8
    mean_among_movers: 4.6
  agent_readiness_lift:
    median: 0.0
    p75: 0.0
    p90: 0.0
    max: 0.0
    mean_among_movers: 0.0
  facet_lift_median_among_movers:
    operational_transparency: 15.8
    developer_ergonomics: 16.6
  composite_band_moves:
    thin -> developing: 1736
    developing -> strong: 655
    emerging -> thin: 367
    strong -> exemplar: 121
    minimal -> emerging: 3
  agent_readiness_band_moves: {}
  method: >-
    each provider's own kin/checks file, the vendor's maps at their stated credit, the scorer's composite
    formula; from -> to, nothing written