> ## Documentation Index > Fetch the complete documentation index at: https://rankspot.ai/docs/llms.txt > Use this file to discover all available pages before exploring further. # List fanout queries > The searches the AI platforms ran while answering your prompts, most-run first. Where a citation tells you which page an engine trusted, a fanout query tells you what it decided it needed to know, so a query you rank for nowhere is a content gap stated in the engine's own words. Defaults to the worklist: queries with no article planned and none written. State is derived from the work linked to the query, the same rule keywords use, so it moves on its own as actions and articles appear. Pass `type=planned`, `type=processed`, `type=all` or `type=archived` for the others. Covers the full history unless you narrow it with `startDate` and `endDate`, which lists only the queries engines ran inside that period and counts `searches` over it; filter by `query` substring to look at one theme. Coverage is not uniform across platforms: not every engine reports its fanout. ## OpenAPI ````yaml https://api.rankspot.ai/docs-json get /v1/ai-visibility/fanouts openapi: 3.0.0 info: title: RankSpot Public API description: >- The RankSpot Public API gives you programmatic access to your workspace data — keywords, backlinks, competitors, planned actions, articles, and more. ## Authentication All endpoints require a Bearer API key in the `Authorization` header: ``` Authorization: Bearer ``` Generate API keys from **Settings → API Keys** in the RankSpot dashboard. Each key is scoped to a single workspace. ## Pagination List endpoints return a consistent envelope: ```json { "data": { "total": 120, "offset": 0, "limit": 20, "count": 20, "items": [...] } } ``` Use `offset` and `limit` query parameters to paginate. Default limit is **20**, maximum is **100**. ## Status codes | Code | Meaning | |------|---------| | 200 | Success | | 201 | Created | | 204 | No content (delete / unarchive) | | 400 | Validation error or business rule violation | | 401 | Missing or invalid API key | | 404 | Resource not found or belongs to another workspace | version: '1.0' contact: {} servers: - url: https://api.rankspot.ai description: Production security: [] tags: - name: AI Visibility description: >- How AI answer engines see your brand. Track prompts, read the answers ChatGPT, Perplexity and Google gave, and work through the pages they cited. Prompts are run on a daily schedule by RankSpot, so answers appear after the next run rather than on demand. - name: Articles description: >- AI-generated blog articles. Articles are created automatically when a `write_article` action is sent to generation — they cannot be created directly via the API. - name: Backlinks description: >- Backlinks discovered for your domain and your competitors' domains. Backlinks are auto-synced on a background interval and cannot be created via the API. - name: Categories description: Categories for organising actions and articles within your workspace. - name: Competitors description: >- Competitor domains tracked in your workspace. Once added, keywords and backlinks for each competitor are fetched asynchronously on a background sync interval. - name: Forum Opportunities description: >- Forum threads and community posts surfaced as potential link-building or engagement opportunities. - name: Keywords description: >- SEO keywords tracked in your workspace, enriched with search volume, competition, and AI-generated relevance scores. - name: People Also Ask description: >- "People also ask" questions discovered from search results for your tracked keywords. - name: Actions description: >- The SEO work planned for your workspace. Each action carries a `type` saying what kind of work it is and which target it points at — see `POST /v1/actions` for the full list. Only `write_article` can be turned into an article; the rest are a person's job. Create actions yourself or let RankSpot generate them. - name: Research description: >- Live lookups against the open web: a Google search and a page fetch. The only endpoints here that spend credits per call. Neither response reports what it cost — the charge is real, and your remaining balance is visible in the RankSpot dashboard. Returns 402 when the workspace is out of credits. - name: Search Console description: >- Google Search Console performance data — clicks, impressions, CTR, and average position for your connected property. Requires GSC to be connected from the RankSpot dashboard. - name: Workspace description: >- Who the workspace is: brand, domain, what the business does and who it sells to. Read this first — the rest of the API returns opportunities that only mean something in the context of a particular business. paths: /v1/ai-visibility/fanouts: get: tags: - AI Visibility summary: List fanout queries description: >- The searches the AI platforms ran while answering your prompts, most-run first. Where a citation tells you which page an engine trusted, a fanout query tells you what it decided it needed to know, so a query you rank for nowhere is a content gap stated in the engine's own words. Defaults to the worklist: queries with no article planned and none written. State is derived from the work linked to the query, the same rule keywords use, so it moves on its own as actions and articles appear. Pass `type=planned`, `type=processed`, `type=all` or `type=archived` for the others. Covers the full history unless you narrow it with `startDate` and `endDate`, which lists only the queries engines ran inside that period and counts `searches` over it; filter by `query` substring to look at one theme. Coverage is not uniform across platforms: not every engine reports its fanout. operationId: AiFanoutQueriesController_findAll parameters: - name: offset required: false in: query schema: minimum: 0 default: 0 type: number - name: limit required: false in: query schema: minimum: 1 maximum: 100 default: 20 type: number - name: startDate required: false in: query description: >- Inclusive start of the period, `YYYY-MM-DD` (UTC). Omit for no lower bound. schema: type: string example: '2026-07-01' - name: endDate required: false in: query description: >- Inclusive end of the period, `YYYY-MM-DD` (UTC). Omit for no upper bound. schema: type: string example: '2026-07-28' - name: query required: false in: query description: >- Filter by query substring, case-insensitive. Stored queries are already lower-cased, so casing here does not matter either. schema: example: screen recorder type: string - name: type required: false in: query description: >- Derived from the work linked to the query, the same rule keywords use. new: the worklist, no article planned and none written (default) | planned: has a `write_article` action but no article yet | processed: has a linked article | all: every non-archived query | archived: queries hidden from the worklist schema: default: new type: string enum: - all - planned - processed - new - archived - name: sortBy required: false in: query schema: default: searches type: string enum: - searches - lastSeenAt - firstSeenAt - query - name: sortOrder required: false in: query schema: default: desc type: string enum: - asc - desc responses: '200': description: '' content: application/json: schema: properties: data: type: object properties: total: type: number example: 100 offset: type: number example: 0 limit: type: number example: 100 count: type: number example: 10 items: type: array items: $ref: '#/components/schemas/AiFanoutQueryDto' security: - bearer: [] components: schemas: AiFanoutQueryDto: type: object properties: id: type: string example: 3f1c2b4e-8d5a-4a2f-9c3e-7b1d6a5e4c20 query: type: string example: best screen recording software for mac 2026 description: >- The search the engine ran. Normalised on write — trimmed, inner whitespace collapsed, lower-cased — so the same search in two casings is one row rather than two. actionIds: example: - 3f1c2b4e-8d5a-4a2f-9c3e-7b1d6a5e4c20 description: >- The `write_article` actions planned for this search. Non-empty with an empty `articleIds` is what `type=planned` selects. type: array items: type: string articleIds: example: - 3f1c2b4e-8d5a-4a2f-9c3e-7b1d6a5e4c20 description: >- The articles written for this search. Non-empty is what `type=processed` selects, whatever the actions say. type: array items: type: string firstSeenAt: format: date-time type: string description: >- When this query was first and last run, ever. Not clipped to the requested period. lastSeenAt: format: date-time type: string searches: type: number example: 14 description: >- Responses in the period whose engine ran this query. An engine runs a query at most once per answer, so this is both the fanout count and the number of answers behind it. platforms: type: array description: Platforms that ran this query in the period. items: $ref: '#/components/schemas/AiPlatform' required: - id - query - actionIds - articleIds - firstSeenAt - lastSeenAt - searches - platforms AiPlatform: type: string enum: - chatgpt - perplexity - google_ai_overview - google_ai_mode securitySchemes: bearer: scheme: bearer bearerFormat: JWT type: http ````