generated: '2026-08-31'
method: searched
source: https://accuracite.com/api-docs
note: >-
  Derived from the response field tables published in the API reference, not from an
  OpenAPI — AccuraCite publishes no machine-readable spec. Every field below is named
  verbatim in the docs.
entities:
- name: VerificationResult
  returned_by: POST /api/v1/verify
  description: Outcome of checking one citation string against the academic indexes.
  fields:
  - name: verification_status
    type: string
    enum: [verified, mismatch, not_found]
    required: true
  - name: mismatches
    type: string[]
    description: >-
      Present only when verification_status is mismatch. Any of: title, author, year,
      DOI, journal, volume, issue, page.
  - name: reason
    type: string|null
    description: Set for some not_found results (e.g. a dead URL); otherwise null.
  - name: source
    type: string|null
    description: >-
      Which index supplied `citation`. One of OpenAlex, Crossref, Semantic Scholar,
      PubMed, DBLP, arXiv, CORE, Google Scholar, "Publisher Metadata", or
      "URL Direct Check" (the last two mean a URL was fetched directly instead of
      matched against a bibliographic record).
  - name: citation
    type: Citation|null
- name: BatchVerificationResult
  returned_by: POST /api/v1/verify (when `texts` is used)
  description: Wrapper returned for a batch request.
  fields:
  - name: results
    type: VerificationResult[]
    description: One per input citation, in the same order as the input array.
- name: GenerationResult
  returned_by: POST /api/v1/generate
  description: Real papers whose abstracts support a supplied claim.
  fields:
  - name: citations
    type: Citation[]
    description: >-
      Up to `limit` entries; may be shorter or empty when fewer qualifying papers exist.
- name: Citation
  description: A bibliographic record, shared by both endpoints.
  fields:
  - {name: title, type: string}
  - {name: author, type: string}
  - {name: doi, type: string|null}
  - {name: url, type: string|null}
  - {name: year, type: string|null}
  - {name: bibtex_entry_type, type: string}
  - {name: volume, type: string|null}
  - {name: issue, type: string|null}
  - {name: page, type: string|null}
  - {name: publisher, type: string|null}
  - {name: container_title, type: string|null}
  - {name: abstract, type: string|null}
relationships:
- from: VerificationResult
  to: Citation
  kind: has_one
  via: citation
  nullable: true
  caveat: >-
    A populated `citation` does NOT imply a match — on not_found it is the closest
    candidate. Branch on verification_status.
- from: BatchVerificationResult
  to: VerificationResult
  kind: has_many
  via: results
- from: GenerationResult
  to: Citation
  kind: has_many
  via: citations
identifiers:
- name: doi
  scheme: DOI (ISO 26324)
  note: The primary cross-system identifier the API emits and consumes.
- name: url
  scheme: URL
  note: >-
    Accepted as INPUT too — a bare URL may be passed as the citation, in which case
    AccuraCite fetches the page and confirms the title/author text appears on it rather
    than matching a bibliographic record.
- name: bibtex_entry_type
  scheme: BibTeX entry type
no_id_prefixes: true
no_persistent_resources: >-
  The API returns no AccuraCite-owned identifiers. There are no accounts, jobs,
  reports or documents addressable over the REST API — every call is stateless and
  returns third-party bibliographic data. There is nothing to GET back later.