generated: '2026-09-10'
method: derived
source: >-
  Derived from the component schemas in openapi/unicorn-screener-openapi.json. The API
  exposes a small, mostly flat model; the load-bearing link is the slug that carries a
  memo request through to its screening status.
entities:
- name: LookupResult
  schema: LookupFound / LookupMissing
  key: name (exact, slug-normalized)
  fields: [found, name, score, classification, summary, website, hq, analyzedAt]
- name: Suggestion
  schema: Suggestions.results[]
  fields: [name, domain, logo]
- name: MemoRequest
  schema: MemoRequest
  fields: [startupName, email, fingerprint, newsletter]
- name: MemoAcceptance
  schema: MemoAccepted
  fields: [success, message, slug, memoUrl]
- name: ScreeningStatus
  schema: ScreeningStatus / UnknownStatus
  key: slug
  fields: [state, slug, publishedSlug, publishable, score, reason, name, startedAt, step, message, done, facts, logoUrl, website]
relationships:
- from: MemoAcceptance
  to: ScreeningStatus
  type: has_one
  via: slug
  detail: The slug returned by requestStartupMemo is polled via getScreeningStatus.
- from: ScreeningStatus
  to: ScreeningStatus
  type: has_one
  via: publishedSlug
  detail: A "moved" state supplies publishedSlug pointing at the canonical published memo status.
- from: ScreeningStatus
  to: LookupResult
  type: belongs_to
  via: name
  detail: A completed screening becomes a lookupable existing result, retrievable by name.