LLM Pulse Answers API
The AI responses behind every metric, with their full text, mentions, citations and sentiment analysis.
The AI responses behind every metric, with their full text, mentions, citations and sentiment analysis.
Every API here is available over the APIs.io API and to AI agents over MCP.
One button, every client — Claude, Cursor, VS Code and the rest.
https://apis.io/mcp
find_apisBrowse and filter every API in the catalog.get_api_artifactsOne API's artifacts, grouped by type.get_openapiThe primary OpenAPI for this API.find_similar_apisAPIs that look like this one.apis_io_searchSTART HERE — APIs, providers and tags for one query, each with its total.resolveTurn a domain, URL or GitHub org into the provider it belongs to.find_cohortsEvery scored population of providers in the catalog.curl "https://apis.io/api/v1/apis/llmpulse-ai:llmpulse-ai-answers-api"
curl "https://apis.io/api/v1/apis?limit=25"
Calls without X-API-Key need no key. Calls with it need a paid plan — get a key.
Free tier, no form to fill in. Signing in shares your email address with us — we store it to create your key and to recognise you if you sign in with another provider. By signing up you are subscribing to the APIs.io and API Evangelist newsletters — you can unsubscribe from any issue. See our Privacy Policy and Terms.
A second provider on the same verified email joins the account you already have.
openapi: 3.2.0
info:
title: LLM Pulse Answers API
description: 'REST API for the LLM Pulse AI visibility platform. Track brand mentions, citations, sentiment, and Share of Voice across AI-generated answers (ChatGPT, Perplexity, Gemini, Google AI Overviews, Google AI Mode, and optional add-on models). Includes the MCP (Model Context Protocol) endpoint for AI clients. MCP is available on every plan and during eligible trials via OAuth 2.1, with Bearer-token API keys also accepted for headless integrations on the Scale plan or above. Full machine-readable reference at https://llmpulse.ai/api-docs/llms-full.txt. An API key can be limited to one or several projects when it is created, and the set never changes. Such a key sees only those projects: any other project id, domain or name answers ERR_PROJECT_NOT_FOUND exactly like a project that does not exist, listings return only those projects'' rows, and operations on the whole account (creating or deleting a project, the project draft wizard) answer ERR_KEY_PROJECT_SCOPED. Operations carry vendor extensions: x-plan is the plan or entitlement the endpoint requires beyond an API key (growth, scale or reputation_access), x-write marks endpoints that need a read and write key, x-mcp-tool names the MCP tool that returns the same data, x-mcp-plan gives that tool''s plan where it differs, and x-rest-only explains an endpoint with no MCP tool. Parameters are read strictly, on REST and MCP alike: a value outside a documented enum, a boolean other than true, false, 1 or 0, an id or count that is not a plain positive number (12abc, 1.0 or 0), an id of a record the project does not have, or a from later than to answers 422 ERR_INVALID_PARAM naming the parameter, never an ignored filter. A listed value, in a query or a request body, matches in any letter case and is read in its documented spelling (granularity=Week reads as week, aggregation_type=BYPAGE as byPage, event_type=Mention.Created as mention.created). page and per_page keep their own rule: a value that is not a number reads as absent and an out-of-range one is clamped.'
version: 1.97.0
contact:
name: LLM Pulse Support
url: https://llmpulse.ai
email: info@llmpulse.ai
license:
name: Proprietary
url: https://llmpulse.ai/terms
servers:
- url: https://api.llmpulse.ai/api/v1
description: Production API
security:
- BearerAuth: []
tags:
- name: Answers
description: The AI responses behind every metric, with their full text, mentions, citations and sentiment analysis.
paths:
/answers:
get:
tags:
- Answers
summary: List AI responses
x-llms-description: Paginated AI responses with truncated content (10k chars), counts, executed_at, no_result and no_result_reason (declined or no_answer). query = full-text search (total becomes the exact match count; rows return snippet + match_count).
operationId: listAnswers
x-mcp-tool: list_answers
description: 'Successful prompt-execution responses with truncated content (max 10,000 chars). Pass `query` for case-insensitive full-text search inside response texts: `total` becomes the exact count of matching responses and each item returns `snippet` + `match_count` instead of `response`/`response_truncated`.'
parameters:
- $ref: '#/components/parameters/ProjectId'
- $ref: '#/components/parameters/Model'
- $ref: '#/components/parameters/CollectionId'
- $ref: '#/components/parameters/CountryCode'
- $ref: '#/components/parameters/LanguageCode'
- $ref: '#/components/parameters/Prompt'
- $ref: '#/components/parameters/PromptType'
- $ref: '#/components/parameters/BrandKind'
- $ref: '#/components/parameters/MentionFilter'
- $ref: '#/components/parameters/CitationFilter'
- $ref: '#/components/parameters/Competitors'
- $ref: '#/components/parameters/ListingFrom'
- $ref: '#/components/parameters/ListingTo'
- $ref: '#/components/parameters/Page'
- $ref: '#/components/parameters/PerPage'
- name: query
in: query
required: false
schema:
type: string
description: Case-insensitive full-text search inside AI response texts. Switches items to snippet + match_count mode.
- name: no_result
in: query
required: false
schema:
type: boolean
description: Filter sentinel non-answers (no answer could be collected after all retries, or the model declined to answer, today only Gemini; excluded from platform metrics). false = only real answers, true = only sentinels, omit = both. Every item carries its own no_result flag, and a sentinel also carries no_result_reason (declined or no_answer).
responses:
'200':
description: Paginated answers. Every item carries app_url, the link that opens the answer in the app
'422':
$ref: '#/components/responses/UnprocessableEntity'
/answers/{id}:
get:
tags:
- Answers
summary: Get one AI response
x-llms-description: Full AI response with mentions, citations, sentiments, sources, shopping_products, brand_entities, fan_out_queries.
operationId: getAnswer
x-mcp-tool: get_answer
description: Full answer with mentions, citations, sentiments, sources, shopping_products, brand_entities, fan_out_queries. `sentiments` need the Growth plan or above and `shopping_products` the Scale plan or above; below those plans each is left out and listed in `plan_required_fields`, which is always present and empty when nothing is left out. Pass `include_source_page_details=true` to nest page-cache metadata under each source.
parameters:
- $ref: '#/components/parameters/ProjectId'
- name: id
in: path
required: true
schema:
type: integer
- name: include_source_page_details
in: query
schema:
type: boolean
default: false
responses:
'200':
description: Answer details
content:
application/json:
schema:
$ref: '#/components/schemas/AnswerDetails'
'404':
$ref: '#/components/responses/NotFound'
components:
parameters:
PerPage:
name: per_page
in: query
schema:
type: integer
default: 20
maximum: 100
minimum: 1
ListingFrom:
name: from
in: query
schema:
type: string
format: date-time
description: Start of the window. Sent alone, the window runs from it to the end of today. A from later than to, or a window longer than 1,500 days, returns 422 ERR_INVALID_PARAM.
Prompt:
name: prompt
in: query
schema:
type: integer
minimum: 1
description: Filter by prompt ID. An id that is not a plain positive number, or that names no prompt of the project, returns 422 ERR_INVALID_PARAM.
ListingTo:
name: to
in: query
schema:
type: string
format: date-time
description: End of the window. A date-only value such as 2026-09-01 covers that whole day. Pass a full timestamp to end the window earlier. Sent without from, the window covers the 30 days that end on it.
PromptType:
name: prompt_type
in: query
schema:
type: string
description: 'One prompt type or a comma-separated list: informational, navigational, commercial, transactional'
CountryCode:
name: country_code
in: query
schema:
type: string
description: One ISO country code or a comma-separated list (e.g. US,GB,DE)
LanguageCode:
name: language_code
in: query
schema:
type: string
description: One ISO language code or a comma-separated list (e.g. en,es,de)
CollectionId:
name: collection_id
in: query
schema:
type: string
pattern: ^\d+(,\d+)*$
example: 12,34
description: 'One collection/tag ID or a comma-separated list of IDs. A query value is always a string on the wire, so it is typed as one: the previous integer-or-string union made generators emit a wrapper type they could not serialize.'
CitationFilter:
name: citation_filter
in: query
schema:
type: string
enum:
- cites_you
- not_cites_you
- cites_competitor
- not_cites_competitor
- you_and_competitor
- competitor_not_you
- you_not_competitor
- cites_no_brands
- no_citations
description: Same two-axis matrix applied to the domains cited in the answer instead of the brands named in it (cites_no_brands still allows other domains), plus no_citations, finished answers that cite no source at all. Independent of mention_filter; pass both to intersect them (e.g. mentions_you + not_cites_you finds answers that talk about you without linking to you).
MentionFilter:
name: mention_filter
in: query
schema:
type: string
enum:
- mentions_you
- not_mentions_you
- mentions_competitor
- not_mentions_competitor
- you_and_competitor
- competitor_not_you
- you_not_competitor
- no_brands
description: 'Filter by which brands are mentioned, as a two-axis matrix (your brand x competitors): mentions_you / not_mentions_you, mentions_competitor / not_mentions_competitor, and the four combined cells you_and_competitor, competitor_not_you (a rival wins and you are absent), you_not_competitor, no_brands (no tracked brand appears, i.e. open space; on answers, finished answers only). Combine with ''competitors'' to narrow the competitor side to specific rivals; on a negative cell that reads ''none of these''. On /dimensions/sources it applies to the crawled content of each cited page instead of the answer text. The legacy value ''competitors_only'' is still accepted as an alias of competitor_not_you.'
BrandKind:
name: brand_kind
in: query
schema:
type: string
enum:
- brand
- brand_other
- non_brand
description: 'Filter by brand kind: brand (own brand/products), brand_other (competitors/other brands), non_brand (generic, no brand named). For fair 1:1 brand-vs-competitor comparisons (visibility, share of voice), use non_brand: brand-focused prompts skew results toward the brand they name. The in-app Overview page applies non_brand by default.'
ProjectId:
name: project_id
in: query
required: true
schema:
type: integer
description: Project ID
Page:
name: page
in: query
schema:
type: integer
default: 1
minimum: 1
Model:
name: model
in: query
schema:
type: string
enum:
- chatgpt
- perplexity
- ai_mode
- ai_overview
- gemini
- copilot
- amazon_rufus
- claude
- grok
- deepseek
- naver_ai
- baidu_ai
- meta_ai
- naver_ai_tab
- qwen
- doubao
- bing_ai_mode
description: 'Filter by one AI model. It must be a model enabled for the API key''s user: an unknown value, or a model the user has not enabled, returns 422 ERR_INVALID_PARAM with the valid values in the message.'
Competitors:
name: competitors
in: query
schema:
type: string
description: Comma-separated competitor IDs (a malformed or unknown ID returns ERR_INVALID_PARAM)
responses:
UnprocessableEntity:
description: Invalid parameters
content:
application/json:
schema:
$ref: '#/components/schemas/Error'
example:
error:
code: ERR_INVALID_PARAM
message: invalid metrics
request_id: abc-123
NotFound:
description: Resource not found
content:
application/json:
schema:
$ref: '#/components/schemas/Error'
example:
error:
code: ERR_PROJECT_NOT_FOUND
message: project_not_found
request_id: abc-123
schemas:
Error:
type: object
properties:
error:
type: object
properties:
code:
type: string
enum:
- ERR_MISSING_AUTH
- ERR_INVALID_API_KEY
- ERR_REVOKED_API_KEY
- ERR_INSUFFICIENT_SCOPE
- ERR_INSUFFICIENT_PERMISSION
- ERR_KEY_PROJECT_SCOPED
- ERR_PLAN_REQUIRED
- ERR_INTEGRATION_UNAVAILABLE
- ERR_GENERATION_FAILED
- ERR_ACCOUNT_INACTIVE
- ERR_DRAFT_NOT_FOUND
- ERR_DRAFT_STATE
- ERR_PROJECT_NOT_FOUND
- ERR_NOT_FOUND
- ERR_SEARCH_CONSOLE_NOT_CONNECTED
- ERR_SEARCH_CONSOLE_ACCESS_REVOKED
- ERR_SEARCH_CONSOLE_UPSTREAM
- ERR_CMS_NOT_CONNECTED
- ERR_CMS_ACCESS_REVOKED
- ERR_CMS_REJECTED
- ERR_CMS_UPSTREAM
- ERR_AI_TRAFFIC_NOT_CONNECTED
- ERR_WEB_ANALYTICS_NOT_CONNECTED
- ERR_WEB_ANALYTICS_INVALID_QUERY
- ERR_WEB_ANALYTICS_ACCESS_REVOKED
- ERR_WEB_ANALYTICS_UPSTREAM
- ERR_AGENT_TRAFFIC_NOT_CONNECTED
- ERR_INVALID_PARAM
- ERR_INVALID_RANGE
- ERR_LIMIT_REACHED
- ERR_QUOTA_EXCEEDED
- ERR_STALE_CONTENT
- ERR_RATE_LIMITED
message:
type: string
meta:
type: object
description: Extra detail for some codes. A not-connected or access-revoked error from the Search Console, AI traffic, web analytics or agent traffic endpoints carries connect_url, the in-app page where the user connects or reconnects the source. A prompt write refused because a launched pitch project's prompts are fixed carries reason pitch_prompts_frozen.
request_id:
type: string
AnswerDetails:
type: object
required:
- plan_required_fields
properties:
id:
type: integer
prompt_id:
type: integer
prompt_text:
type: string
model:
type: string
response:
type:
- string
- 'null'
description: The AI answer. For a declined answer (no_result_reason declined) it is the reply the model gave instead; for no_answer it is a placeholder, not an AI response
response_truncated:
type: boolean
executed_at:
type:
- string
- 'null'
format: date-time
duration_ms:
type:
- number
- 'null'
description: Milliseconds, rounded to one decimal place
success:
type:
- boolean
- 'null'
description: Null while the answer is still pending
no_result:
type: boolean
description: True for a sentinel non-answer (no answer could be collected after all retries, or the model declined to answer, today only Gemini); excluded from platform metrics
no_result_reason:
type: string
enum:
- declined
- no_answer
description: Present only when no_result is true. declined = the model declined to answer (today only Gemini, shown as Declined in the app) and response holds its reply; no_answer = no answer could be collected
fan_out_queries:
type:
- array
- 'null'
items:
type: string
mentions:
type: array
items:
type: object
citations:
type: array
items:
type: object
competitor_mentions:
type: array
items:
type: object
competitor_citations:
type: array
items:
type: object
sentiments:
type: array
description: Needs the Growth plan or above; left out below it and named in plan_required_fields
items:
type: object
sources:
type: array
items:
type: object
shopping_products:
type: array
description: Needs the Scale plan or above; left out below it and named in plan_required_fields
items:
type: object
brand_entities:
type: array
items:
type: object
local_businesses:
type: array
items:
type: object
locale:
type: object
properties:
country_code:
type:
- string
- 'null'
language_code:
type:
- string
- 'null'
plan_required_fields:
type: array
description: The fields left out of this answer because the account's plan does not include them (sentiments below Growth, shopping_products below Scale). Always present, empty when nothing is left out
items:
type: string
app_url:
type: string
format: uri
description: Opens this answer in the app. The link names its project, so it opens there for any user with access to that project
request_id:
type: string
securitySchemes:
BearerAuth:
type: http
scheme: bearer
description: API key authentication. Keys are managed at https://llmpulse.ai/app/api_keys (Scale plan or above). Used for the REST API and as a headless option for /mcp. A key can be limited to one or several projects when it is created; the set cannot be changed later.
OAuth2:
type: oauth2
description: OAuth 2.1 with PKCE-S256 and Dynamic Client Registration. Accepted on /mcp only, available on every plan and during eligible trials. Discovery metadata at /.well-known/oauth-authorization-server.
flows:
authorizationCode:
authorizationUrl: https://api.llmpulse.ai/oauth/authorize
tokenUrl: https://api.llmpulse.ai/oauth/token
scopes:
mcp:read: Read data via MCP tools
mcp:write: Create resources via MCP write tools