Fontes e inteligência de citações

Todos os URLs citados pelos modelos, além de vistas agrupadas por URL, domínio e host, metadados da cache de página, evidência de menções dentro das páginas citadas, listas de ocorrências e conteúdo em cache sanitizado.

Os URLs citados mantêm as suas métricas de citação quando o conteúdo da página não está disponível. As menções de página desconhecidas passam a devolver null; content_gap_status inclui content_unavailable. Os resultados de rastreio vazios já não aparecem como páginas analisadas. Os estados 404 e 410 observados continuam visíveis.

GET`/dimensions/sources`

Lista as fontes não rejeitadas produzidas pelas execuções de prompts.

Parâmetros:

project_idobrigatório
model
collection_id
promptID do prompt
country_code
language_code
from
to
page
per_page
source_typeowned|competitor|third_party
mention_filter
competitors

Qualquer outro valor de source_type, incluindo um que só difira nas maiúsculas e minúsculas, devolve ERR_INVALID_PARAM. Aqui, mention_filter aplica-se às marcas mencionadas no conteúdo recolhido de cada página citada, não na resposta da IA, à semelhança da página Citações.

GET`/citation_intelligence/groups`

Inteligência de citações agrupada por url, domain ou host, com desagregação por modelo, taxa de citação e posição média de citação (ignora as linhas com position = 0).

Parâmetros:

project_idobrigatório
viewurl|domain|host
page
per_page
ordergroup_key, total_responses, total_citations, citation_rate, avg_citation_position, first_seen_at, last_seen_at
direction
model
collection_id
country_code
language_code
prompt
from
to
query
source_typeowned|competitor|third_party|social_media|own_domain|ugc|background
sentimentnegative
content_gapmentioned|gap

Valores de enum inválidos devolvem ERR_INVALID_PARAM.

brand_kind (brand|brand_other|non_brand) mantém apenas as respostas a prompts com esse foco de marca. key devolve só o grupo indicado da vista escolhida: um domínio, um host ou um URL citado, tal como aparece na lista. A correspondência de domínios e hosts não distingue maiúsculas de minúsculas e ignora www.; se não houver correspondência, é devolvida uma lista vazia, e não um erro. Com view=url, o valor indicado corresponde a um único URL citado em bruto, pelo que as variantes desse URL continuam em linhas separadas; GET /citation_intelligence/urls/:url_sha256 agrega-as, tal como faz a aplicação. Com key, include_series=true acrescenta citation_series à linha, no formato que GET /citation_intelligence/urls/:url_sha256 devolve, e conta cada resposta apenas uma vez, mesmo que cite várias páginas do grupo. include_series sem key devolve ERR_INVALID_PARAM.

GET`/citation_intelligence/mentions_by_domain`

Para as respostas em que cada domínio de origem indicado é citado, a percentagem dessas respostas que menciona a sua marca em comparação com cada concorrente. A marca e os concorrentes são normalizados para somar 100% por domínio, refletindo a definição de menções do Share of Voice. Indique vários domínios para analisar toda a matriz numa só chamada.

Parâmetros:

project_idobrigatório
domainsarray obrigatório, por exemplo domains[]=example.com, máx. 50
model
collection_id
collection_ids
country_code
language_code
prompt
brand_kindbrand|brand_other|non_brand
from
to

Um domains vazio devolve ERR_INVALID_PARAM. Cada linha tem responses_citing, total_mentions e um array actors (marca e concorrentes com mentions e share).

GET`/citation_intelligence/urls/:url_sha256`

Detalhe ao nível do URL para uma página citada: totais, taxa de citação, posição média de citação, contagens por tipo de origem, contagens por modelo, variantes de URL distintas, metadados da cache de página e evidências de menções na página com fragmentos e etiquetas de intervenientes.

Parâmetros: project_id (obrigatório), :url_sha256 (64 caracteres hexadecimais no URL), os mesmos filtros opcionais que /citation_intelligence/groups. ERR_NOT_FOUND apenas quando o URL nunca foi citado pelo projeto; os filtros que esvaziam o resultado continuam a devolver 200 com estatísticas a zero.

include_series=true acrescenta citation_series, as respostas que citam o URL com os mesmos filtros (key não se aplica aqui): total (igual a stats.total_responses), by_model (respostas por modelo de IA, por ordem decrescente; pode incluir modelos ativados além dos sete que model aceita) e data, um ponto { date, value } por cada período de granularity. O período é day, week ou month, de acordo com a frequência de monitorização da conta, e uma conta diária recebe semanas, a começar à segunda-feira, quando as respostas filtradas abrangem mais de 60 dias. Cada resposta é datada pelo respetivo executed_at no fuso horário Europe/Madrid, a coluna pela qual from e to filtram. Um período com respostas que não citaram o URL apresenta 0. Sem include_series, a resposta do endpoint não muda.

GET`/citation_intelligence/urls/:url_sha256/occurrences`

Ocorrências paginadas de um URL citado em execuções de prompts, com excerto da resposta, texto do prompt, modelo, executed_at, citation_position e source_type por linha.

Parâmetros: project_id (obrigatório), :url_sha256 (64 caracteres hexadecimais), page, per_page, os mesmos filtros opcionais que /citation_intelligence/groups.

GET`/citation_intelligence/urls/:url_sha256/content`

Conteúdo de página em cache sanitizado para um URL citado: metadados da cache de página, evidências e fragmentos de menções, texto simples integral e HTML renderizado (scripts e marcação não segura removidos).

Parâmetros:

project_idobrigatório
:url_sha25664 caracteres hexadecimais

Exemplos de código

Download: OpenAPI Swagger Postman

Perguntar à IA: Abrir no ChatGPT Abrir no Claude Ver como Markdown

Linguagem

Request

curl -X GET -H "Authorization: Bearer YOUR_API_KEY" \
  "https://api.llmpulse.ai/api/v1/dimensions/sources?project_id=PROJECT_ID"

const response = await fetch("https://api.llmpulse.ai/api/v1/dimensions/sources?project_id=PROJECT_ID", {
  headers: { "Authorization": "Bearer YOUR_API_KEY" }
});
const data = await response.json();

import requests

response = requests.get(
    "https://api.llmpulse.ai/api/v1/dimensions/sources?project_id=PROJECT_ID",
    headers={"Authorization": "Bearer YOUR_API_KEY"}
)
data = response.json()

require 'httparty'

response = HTTParty.get(
  "https://api.llmpulse.ai/api/v1/dimensions/sources?project_id=PROJECT_ID",
  headers: { "Authorization" => "Bearer YOUR_API_KEY" }
)
data = response.parsed_response

req, _ := http.NewRequest("GET", "https://api.llmpulse.ai/api/v1/dimensions/sources?project_id=PROJECT_ID", nil)
req.Header.Set("Authorization", "Bearer YOUR_API_KEY")
client := &http.Client{}
resp, _ := client.Do(req)
defer resp.Body.Close()

$client = new GuzzleHttp\Client();
$response = $client->request('GET', 'https://api.llmpulse.ai/api/v1/dimensions/sources?project_id=PROJECT_ID', [
    'headers' => ['Authorization' => 'Bearer YOUR_API_KEY']
]);
$data = json_decode($response->getBody(), true);

HttpRequest request = HttpRequest.newBuilder()
    .uri(URI.create("https://api.llmpulse.ai/api/v1/dimensions/sources?project_id=PROJECT_ID"))
    .header("Authorization", "Bearer YOUR_API_KEY")
    .GET()
    .build();
HttpResponse<String> response = client.send(request, HttpResponse.BodyHandlers.ofString());

Response

{ "project_id":4, "page":1, "per_page":20, "total":57,
  "data":[ { "id":501, "prompt_execution_id":9001, "domain":"example.com", "url":"https://example.com/x", "position":1 } ],
  "request_id":"..." }

Request

curl -X GET -H "Authorization: Bearer YOUR_API_KEY" \
  "https://api.llmpulse.ai/api/v1/citation_intelligence/groups?project_id=PROJECT_ID&view=domain&range=30"

const response = await fetch("https://api.llmpulse.ai/api/v1/citation_intelligence/groups?project_id=PROJECT_ID&view=domain&range=30", {
  headers: { "Authorization": "Bearer YOUR_API_KEY" }
});
const data = await response.json();

import requests

response = requests.get(
    "https://api.llmpulse.ai/api/v1/citation_intelligence/groups?project_id=PROJECT_ID&view=domain&range=30",
    headers={"Authorization": "Bearer YOUR_API_KEY"}
)
data = response.json()

require 'httparty'

response = HTTParty.get(
  "https://api.llmpulse.ai/api/v1/citation_intelligence/groups?project_id=PROJECT_ID&view=domain&range=30",
  headers: { "Authorization" => "Bearer YOUR_API_KEY" }
)
data = response.parsed_response

req, _ := http.NewRequest("GET", "https://api.llmpulse.ai/api/v1/citation_intelligence/groups?project_id=PROJECT_ID&view=domain&range=30", nil)
req.Header.Set("Authorization", "Bearer YOUR_API_KEY")
client := &http.Client{}
resp, _ := client.Do(req)
defer resp.Body.Close()

$client = new GuzzleHttp\Client();
$response = $client->request('GET', 'https://api.llmpulse.ai/api/v1/citation_intelligence/groups?project_id=PROJECT_ID&view=domain&range=30', [
    'headers' => ['Authorization' => 'Bearer YOUR_API_KEY']
]);
$data = json_decode($response->getBody(), true);

HttpRequest request = HttpRequest.newBuilder()
    .uri(URI.create("https://api.llmpulse.ai/api/v1/citation_intelligence/groups?project_id=PROJECT_ID&view=domain&range=30"))
    .header("Authorization", "Bearer YOUR_API_KEY")
    .GET()
    .build();
HttpResponse<String> response = client.send(request, HttpResponse.BodyHandlers.ofString());

Request

curl -X GET -H "Authorization: Bearer YOUR_API_KEY" \
  "https://api.llmpulse.ai/api/v1/citation_intelligence/mentions_by_domain?project_id=PROJECT_ID&domains=nytimes.com&range=30"

const response = await fetch("https://api.llmpulse.ai/api/v1/citation_intelligence/mentions_by_domain?project_id=PROJECT_ID&domains=nytimes.com&range=30", {
  headers: { "Authorization": "Bearer YOUR_API_KEY" }
});
const data = await response.json();

import requests

response = requests.get(
    "https://api.llmpulse.ai/api/v1/citation_intelligence/mentions_by_domain?project_id=PROJECT_ID&domains=nytimes.com&range=30",
    headers={"Authorization": "Bearer YOUR_API_KEY"}
)
data = response.json()

require 'httparty'

response = HTTParty.get(
  "https://api.llmpulse.ai/api/v1/citation_intelligence/mentions_by_domain?project_id=PROJECT_ID&domains=nytimes.com&range=30",
  headers: { "Authorization" => "Bearer YOUR_API_KEY" }
)
data = response.parsed_response

req, _ := http.NewRequest("GET", "https://api.llmpulse.ai/api/v1/citation_intelligence/mentions_by_domain?project_id=PROJECT_ID&domains=nytimes.com&range=30", nil)
req.Header.Set("Authorization", "Bearer YOUR_API_KEY")
client := &http.Client{}
resp, _ := client.Do(req)
defer resp.Body.Close()

$client = new GuzzleHttp\Client();
$response = $client->request('GET', 'https://api.llmpulse.ai/api/v1/citation_intelligence/mentions_by_domain?project_id=PROJECT_ID&domains=nytimes.com&range=30', [
    'headers' => ['Authorization' => 'Bearer YOUR_API_KEY']
]);
$data = json_decode($response->getBody(), true);

HttpRequest request = HttpRequest.newBuilder()
    .uri(URI.create("https://api.llmpulse.ai/api/v1/citation_intelligence/mentions_by_domain?project_id=PROJECT_ID&domains=nytimes.com&range=30"))
    .header("Authorization", "Bearer YOUR_API_KEY")
    .GET()
    .build();
HttpResponse<String> response = client.send(request, HttpResponse.BodyHandlers.ofString());

Request

curl -X GET -H "Authorization: Bearer YOUR_API_KEY" \
  "https://api.llmpulse.ai/api/v1/citation_intelligence/urls/URL_SHA256?project_id=PROJECT_ID&include_series=true&from=2026-07-01&to=2026-09-30"

const response = await fetch("https://api.llmpulse.ai/api/v1/citation_intelligence/urls/URL_SHA256?project_id=PROJECT_ID&include_series=true&from=2026-07-01&to=2026-09-30", {
  headers: { "Authorization": "Bearer YOUR_API_KEY" }
});
const data = await response.json();

import requests

response = requests.get(
    "https://api.llmpulse.ai/api/v1/citation_intelligence/urls/URL_SHA256?project_id=PROJECT_ID&include_series=true&from=2026-07-01&to=2026-09-30",
    headers={"Authorization": "Bearer YOUR_API_KEY"}
)
data = response.json()

require 'httparty'

response = HTTParty.get(
  "https://api.llmpulse.ai/api/v1/citation_intelligence/urls/URL_SHA256?project_id=PROJECT_ID&include_series=true&from=2026-07-01&to=2026-09-30",
  headers: { "Authorization" => "Bearer YOUR_API_KEY" }
)
data = response.parsed_response

req, _ := http.NewRequest("GET", "https://api.llmpulse.ai/api/v1/citation_intelligence/urls/URL_SHA256?project_id=PROJECT_ID&include_series=true&from=2026-07-01&to=2026-09-30", nil)
req.Header.Set("Authorization", "Bearer YOUR_API_KEY")
client := &http.Client{}
resp, _ := client.Do(req)
defer resp.Body.Close()

$client = new GuzzleHttp\Client();
$response = $client->request('GET', 'https://api.llmpulse.ai/api/v1/citation_intelligence/urls/URL_SHA256?project_id=PROJECT_ID&include_series=true&from=2026-07-01&to=2026-09-30', [
    'headers' => ['Authorization' => 'Bearer YOUR_API_KEY']
]);
$data = json_decode($response->getBody(), true);

HttpRequest request = HttpRequest.newBuilder()
    .uri(URI.create("https://api.llmpulse.ai/api/v1/citation_intelligence/urls/URL_SHA256?project_id=PROJECT_ID&include_series=true&from=2026-07-01&to=2026-09-30"))
    .header("Authorization", "Bearer YOUR_API_KEY")
    .GET()
    .build();
HttpResponse<String> response = client.send(request, HttpResponse.BodyHandlers.ofString());

Request

curl -X GET -H "Authorization: Bearer YOUR_API_KEY" \
  "https://api.llmpulse.ai/api/v1/citation_intelligence/urls/URL_SHA256/occurrences?project_id=PROJECT_ID&per_page=50"

const response = await fetch("https://api.llmpulse.ai/api/v1/citation_intelligence/urls/URL_SHA256/occurrences?project_id=PROJECT_ID&per_page=50", {
  headers: { "Authorization": "Bearer YOUR_API_KEY" }
});
const data = await response.json();

import requests

response = requests.get(
    "https://api.llmpulse.ai/api/v1/citation_intelligence/urls/URL_SHA256/occurrences?project_id=PROJECT_ID&per_page=50",
    headers={"Authorization": "Bearer YOUR_API_KEY"}
)
data = response.json()

require 'httparty'

response = HTTParty.get(
  "https://api.llmpulse.ai/api/v1/citation_intelligence/urls/URL_SHA256/occurrences?project_id=PROJECT_ID&per_page=50",
  headers: { "Authorization" => "Bearer YOUR_API_KEY" }
)
data = response.parsed_response

req, _ := http.NewRequest("GET", "https://api.llmpulse.ai/api/v1/citation_intelligence/urls/URL_SHA256/occurrences?project_id=PROJECT_ID&per_page=50", nil)
req.Header.Set("Authorization", "Bearer YOUR_API_KEY")
client := &http.Client{}
resp, _ := client.Do(req)
defer resp.Body.Close()

$client = new GuzzleHttp\Client();
$response = $client->request('GET', 'https://api.llmpulse.ai/api/v1/citation_intelligence/urls/URL_SHA256/occurrences?project_id=PROJECT_ID&per_page=50', [
    'headers' => ['Authorization' => 'Bearer YOUR_API_KEY']
]);
$data = json_decode($response->getBody(), true);

HttpRequest request = HttpRequest.newBuilder()
    .uri(URI.create("https://api.llmpulse.ai/api/v1/citation_intelligence/urls/URL_SHA256/occurrences?project_id=PROJECT_ID&per_page=50"))
    .header("Authorization", "Bearer YOUR_API_KEY")
    .GET()
    .build();
HttpResponse<String> response = client.send(request, HttpResponse.BodyHandlers.ofString());

Request

curl -X GET -H "Authorization: Bearer YOUR_API_KEY" \
  "https://api.llmpulse.ai/api/v1/citation_intelligence/urls/URL_SHA256/content?project_id=PROJECT_ID"

const response = await fetch("https://api.llmpulse.ai/api/v1/citation_intelligence/urls/URL_SHA256/content?project_id=PROJECT_ID", {
  headers: { "Authorization": "Bearer YOUR_API_KEY" }
});
const data = await response.json();

import requests

response = requests.get(
    "https://api.llmpulse.ai/api/v1/citation_intelligence/urls/URL_SHA256/content?project_id=PROJECT_ID",
    headers={"Authorization": "Bearer YOUR_API_KEY"}
)
data = response.json()

require 'httparty'

response = HTTParty.get(
  "https://api.llmpulse.ai/api/v1/citation_intelligence/urls/URL_SHA256/content?project_id=PROJECT_ID",
  headers: { "Authorization" => "Bearer YOUR_API_KEY" }
)
data = response.parsed_response

req, _ := http.NewRequest("GET", "https://api.llmpulse.ai/api/v1/citation_intelligence/urls/URL_SHA256/content?project_id=PROJECT_ID", nil)
req.Header.Set("Authorization", "Bearer YOUR_API_KEY")
client := &http.Client{}
resp, _ := client.Do(req)
defer resp.Body.Close()

$client = new GuzzleHttp\Client();
$response = $client->request('GET', 'https://api.llmpulse.ai/api/v1/citation_intelligence/urls/URL_SHA256/content?project_id=PROJECT_ID', [
    'headers' => ['Authorization' => 'Bearer YOUR_API_KEY']
]);
$data = json_decode($response->getBody(), true);

HttpRequest request = HttpRequest.newBuilder()
    .uri(URI.create("https://api.llmpulse.ai/api/v1/citation_intelligence/urls/URL_SHA256/content?project_id=PROJECT_ID"))
    .header("Authorization", "Bearer YOUR_API_KEY")
    .GET()
    .build();
HttpResponse<String> response = client.send(request, HttpResponse.BodyHandlers.ofString());

Fontes e inteligência de citações

GET/dimensions/sources

GET/citation_intelligence/groups

GET/citation_intelligence/mentions_by_domain

GET/citation_intelligence/urls/:url_sha256

GET/citation_intelligence/urls/:url_sha256/occurrences

GET/citation_intelligence/urls/:url_sha256/content

GET`/dimensions/sources`

GET`/citation_intelligence/groups`

GET`/citation_intelligence/mentions_by_domain`

GET`/citation_intelligence/urls/:url_sha256`

GET`/citation_intelligence/urls/:url_sha256/occurrences`

GET`/citation_intelligence/urls/:url_sha256/content`