Analyze answer claims and cited sources

Analyzes an answer against its cited pages and returns a version 4 result: the searches the engine ran, the pages it retrieved and cited and who controls each, the brands the answer names and the pages naming them, the brand’s visibility stage and prioritized actions with steps (with brand), the paragraphs of each cited page mapped to UTF-16 answer ranges with explanations, and short summaries of what the pages add up to. Post-hoc correspondence, not causal attribution. The retired analysis.version=1 opt-in is rejected with 400 because insights are automatic. Tasks stored with the older v2 result stay readable as-is. Size is not a failure mode: anything the request contract accepts (an answer up to 200,000 characters with up to 100 citations) is analyzed, with long answers, large source sets and very long pages handled by splitting the work. Depth follows the answer rather than the page, so a large or repetitive page yields its strongest corresponding passages instead of every occurrence, and markup-only pages such as sitemaps or link indexes contribute no links.

POST

citation-analysis

curl --request POST \
  --url https://api.querying.ai/v1/citation-analysis \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "answer": "Teams plan starts at $19.95 per month for up to 10 users. The Business plan adds audit logs.",
  "prompt": "Which plan fits a small team?",
  "citations": [
    {
      "url": "https://example.com/pricing"
    },
    {
      "url": "https://review.example.com/note"
    }
  ]
}
'

{
  "success": true,
  "task": {
    "id": "3c90c3cc-0d44-4b50-8888-8dd25736052a",
    "taskType": "CHATGPT",
    "status": "QUEUED",
    "priority": 123,
    "createdAt": "2023-11-07T05:31:56Z",
    "idempotencyKey": "<string>"
  },
  "credits": {
    "creditsToCharge": 123,
    "creditsCharged": 123
  }
}

{
  "success": false,
  "error": {
    "code": "<string>",
    "message": "<string>",
    "details": [
      {
        "field": "<string>",
        "message": "<string>"
      }
    ],
    "timestamp": "2023-11-07T05:31:56Z"
  }
}

{
  "success": false,
  "error": {
    "code": "<string>",
    "message": "<string>",
    "details": [
      {
        "field": "<string>",
        "message": "<string>"
      }
    ],
    "timestamp": "2023-11-07T05:31:56Z"
  }
}

{
  "success": false,
  "error": {
    "code": "<string>",
    "message": "<string>",
    "details": [
      {
        "field": "<string>",
        "message": "<string>"
      }
    ],
    "timestamp": "2023-11-07T05:31:56Z"
  }
}

{
  "success": false,
  "error": {
    "code": "<string>",
    "message": "<string>",
    "details": [
      {
        "field": "<string>",
        "message": "<string>"
      }
    ],
    "timestamp": "2023-11-07T05:31:56Z"
  }
}

{
  "success": false,
  "error": {
    "code": "<string>",
    "message": "<string>",
    "details": [
      {
        "field": "<string>",
        "message": "<string>"
      }
    ],
    "timestamp": "2023-11-07T05:31:56Z"
  }
}

{
  "success": false,
  "error": {
    "code": "<string>",
    "message": "<string>",
    "details": [
      {
        "field": "<string>",
        "message": "<string>"
      }
    ],
    "timestamp": "2023-11-07T05:31:56Z"
  }
}

{
  "success": false,
  "error": {
    "code": "<string>",
    "message": "<string>",
    "details": [
      {
        "field": "<string>",
        "message": "<string>"
      }
    ],
    "timestamp": "2023-11-07T05:31:56Z"
  }
}

Authorizations

Authorization

string

header

required

Your API key, sent as Authorization: Bearer <key>.

Body

application/json

Option 1
Option 2

answer

string

required

Required unless taskId is sent. The AI answer to analyze, as raw markdown (keep link fragments intact). 1-200,000 characters at submission; the analysis itself has a smaller per-answer processing budget, and an answer above it fails explicitly instead of being truncated.

Required string length: 1 - 200000

citations

object[]

required

Required unless taskId is sent. Cited pages (1-100, at least one) plus optional retrieved-but-not-cited pages (cited: false, at most 50); at most 10 discussion-thread URLs. Total HTTP body <=1 MiB.

Maximum array length: 150

Show child attributes

taskId

string

Instead of answer and citations: the id of one of your COMPLETED answer tasks that is still readable through GET /v1/async/task/{id}. Its answer Markdown (text when there is none) becomes answer, and its cited URLs become citations in the engine's order (first 100, at most 10 discussion threads). Its prompt, country and engine are used unless you send them. Sending taskId together with answer or citations is a 400; an unknown or expired id is a 404. For CHATGPT the task's searches become queries and every retrieved page becomes a citation with cited, position, citedText, publishedAt and query; for Google AI Overview, organic results outside the overview's sources are added as non-cited pages. include.citationAnalysis fills them the same way.

Maximum string length: 128

prompt

string

The original question.

Maximum string length: 200000

country

string

engine

string

The AI engine that produced the answer.

Maximum string length: 64

brand

string

Your brand name. With it, the result adds visibility and actions.

Maximum string length: 200

competitors

string[]

Up to 25 competitor names, reported in brands with role competitor.

Maximum array length: 25

Maximum string length: 200

brandDomains

string[]

Hostnames of your brand's own sites (e.g. coway.com): bare hostnames without scheme or path. Requires brand; sent without it, the request is rejected with 422. Without brandDomains, a host whose name contains the brand name counts as the brand's own site.

Maximum array length: 10

Maximum string length: 253

Pattern: ^[a-zA-Z0-9.-]{1,253}$

brandAliases

string[]

Other names of your brand (product lines, local spellings). Requires brand; sent without it, the request is rejected with 422.

Maximum array length: 10

Maximum string length: 200

queries

string[]

Searches the engine ran for this answer.

Maximum array length: 20

Maximum string length: 500

idempotencyKey

string

priority

integer

webhook

object

Show child attributes

Response

Queued; poll /v1/async/task/{id}. Its response follows CitationAnalysisResult.

success

boolean

required

task

object

required

Show child attributes

credits

object

required

Credits, with up to two decimals (for example 1.25). The submit response always shows 0: a hold is placed on your balance, but nothing is charged yet. Polling shows 0 until the task is terminal, then the settled charge: the task's price for COMPLETED, 0 for FAILED. A task webhook reports the same charge in creditsCharged; on a FAILED task it reports the task's price in creditsToCharge and 0 in creditsCharged. CITATION_ANALYSIS and WEB_CRAWL are charged by usage, up to the hold placed at submit.

Show child attributes

curl --request POST \
  --url https://api.querying.ai/v1/citation-analysis \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "answer": "Teams plan starts at $19.95 per month for up to 10 users. The Business plan adds audit logs.",
  "prompt": "Which plan fits a small team?",
  "citations": [
    {
      "url": "https://example.com/pricing"
    },
    {
      "url": "https://review.example.com/note"
    }
  ]
}
'

{
  "success": true,
  "task": {
    "id": "3c90c3cc-0d44-4b50-8888-8dd25736052a",
    "taskType": "CHATGPT",
    "status": "QUEUED",
    "priority": 123,
    "createdAt": "2023-11-07T05:31:56Z",
    "idempotencyKey": "<string>"
  },
  "credits": {
    "creditsToCharge": 123,
    "creditsCharged": 123
  }
}

{
  "success": false,
  "error": {
    "code": "<string>",
    "message": "<string>",
    "details": [
      {
        "field": "<string>",
        "message": "<string>"
      }
    ],
    "timestamp": "2023-11-07T05:31:56Z"
  }
}

{
  "success": false,
  "error": {
    "code": "<string>",
    "message": "<string>",
    "details": [
      {
        "field": "<string>",
        "message": "<string>"
      }
    ],
    "timestamp": "2023-11-07T05:31:56Z"
  }
}

{
  "success": false,
  "error": {
    "code": "<string>",
    "message": "<string>",
    "details": [
      {
        "field": "<string>",
        "message": "<string>"
      }
    ],
    "timestamp": "2023-11-07T05:31:56Z"
  }
}

{
  "success": false,
  "error": {
    "code": "<string>",
    "message": "<string>",
    "details": [
      {
        "field": "<string>",
        "message": "<string>"
      }
    ],
    "timestamp": "2023-11-07T05:31:56Z"
  }
}

{
  "success": false,
  "error": {
    "code": "<string>",
    "message": "<string>",
    "details": [
      {
        "field": "<string>",
        "message": "<string>"
      }
    ],
    "timestamp": "2023-11-07T05:31:56Z"
  }
}

{
  "success": false,
  "error": {
    "code": "<string>",
    "message": "<string>",
    "details": [
      {
        "field": "<string>",
        "message": "<string>"
      }
    ],
    "timestamp": "2023-11-07T05:31:56Z"
  }
}

{
  "success": false,
  "error": {
    "code": "<string>",
    "message": "<string>",
    "details": [
      {
        "field": "<string>",
        "message": "<string>"
      }
    ],
    "timestamp": "2023-11-07T05:31:56Z"
  }
}

Tasks

Batches

Location grids

Research

AI Rendering

Monitors

Exports

Sub-accounts

Operations

Analyze answer claims and cited sources

Authorizations

Body

Response