> ## Documentation Index > Fetch the complete documentation index at: https://docs.verseodin.com/llms.txt > Use this file to discover all available pages before exploring further. # Get full prompts row(s) — including AI answer + citations > Returns full prompt records — every column documented in `PromptRow`, including `response_text` (the full AI answer body) and the `my_citations` / `competitor_citations` URL arrays. Offset-paginated. Iterate `offset` by `limit` until `data` comes back empty. ## OpenAPI ````yaml /openapi.yaml get /universes/{id}/prompts openapi: 3.1.0 info: title: Verseodin API version: 1.0.0 description: | Read-only HTTP API for pulling Verseodin dashboard data: universes, history rows (per-day aggregates of brand visibility, citations, and mentions across AI search engines), and full prompt records including the AI's response text and cited URLs. The same key authorises the Verseodin Claude / MCP connector — one credential, two surfaces. servers: - url: https://verseodin.com/api/v1 description: Production security: - BearerAuth: [] paths: /universes/{id}/prompts: get: tags: - Prompts summary: Get full prompts row(s) — including AI answer + citations description: | Returns full prompt records — every column documented in `PromptRow`, including `response_text` (the full AI answer body) and the `my_citations` / `competitor_citations` URL arrays. Offset-paginated. Iterate `offset` by `limit` until `data` comes back empty. operationId: listPrompts parameters: - $ref: '#/components/parameters/UniverseId' - $ref: '#/components/parameters/Day' - $ref: '#/components/parameters/Engine' - name: status in: query required: false schema: type: string enum: - pending - processing - completed - failed description: Filter by lifecycle status. - name: limit in: query required: false schema: type: integer minimum: 1 maximum: 1000 default: 200 - name: offset in: query required: false schema: type: integer minimum: 0 default: 0 responses: '200': description: OK content: application/json: schema: type: object properties: data: type: array items: $ref: '#/components/schemas/PromptRow' limit: type: integer offset: type: integer '400': $ref: '#/components/responses/BadRequest' '401': $ref: '#/components/responses/Unauthorized' '403': $ref: '#/components/responses/Forbidden' '404': $ref: '#/components/responses/NotFound' components: parameters: UniverseId: name: id in: path required: true schema: type: string format: uuid description: Universe UUID — see `GET /universes`. Day: name: day in: query required: false schema: type: string format: date description: >- Pin the query to a specific day (`YYYY-MM-DD`). Without this, returns the most recent day. Engine: name: engine in: query required: false schema: type: string enum: - chatgpt - gemini - grok - claude description: Filter to a single AI engine. Omit for all. schemas: PromptRow: type: object description: | Single row from the `prompts` table — one row per `(universe, day, engine, prompt_id)`. Every column is documented below as a separate response field. **`response_text` carries the full AI answer body** and `my_citations` / `competitor_citations` carry the actual cited URLs. properties: id: type: string format: uuid description: Prompt row UUID. day: type: string format: date user_id: type: string format: uuid universe_id: type: string format: uuid job_id: type: string format: uuid prompt_id: type: integer description: >- 1-indexed position of the prompt within the universe (stable across days). prompt_text: type: string description: The natural-language query that was sent to the AI assistant. engine: type: string enum: - chatgpt - gemini - grok - claude description: AI assistant the prompt was run against. engine_account: type: string nullable: true description: >- Scraper pipeline that produced the row. One of `BD_chatgpt`, `BD_gemini`, `BD_grok`, `BD_perplexity` (Bright Data) or null/empty for legacy Cloro rows. website: type: object description: >- Universe's website object captured at run time — `{url, location, brand_tokens?}`. competitor_websites: type: array description: Competitor URL list captured at run time. items: type: object priority_level: type: integer description: Internal scheduling priority (0 = normal, higher = run sooner). provider: type: string nullable: true description: Underlying scraping provider (e.g. `brightdata`). attempts: type: integer nullable: true description: >- How many scrape attempts were spent before the row reached its terminal state. snapshot_id: type: string nullable: true description: >- Bright Data snapshot identifier — useful for cross-referencing in their dashboard. status: type: string enum: - pending - processing - completed - failed description: Terminal states are `completed` and `failed`. error_text: type: string nullable: true description: Failure reason — only set when `status='failed'`. created_at: type: string format: date-time started_at: type: string format: date-time nullable: true description: When the scrape was claimed for processing. finished_at: type: string format: date-time nullable: true description: When the scrape reached a terminal state. response_text: type: string nullable: true description: | **Full AI answer body** as returned by the assistant. Plain text with markdown formatting preserved. Empty/null while `status` is pending or processing. appeared_links: type: array description: >- Every URL the AI mentioned in its answer (in order of appearance, may include duplicates). items: type: string appeared_links_unique: type: array description: De-duplicated `appeared_links`. items: type: string appeared_links_run1: type: array description: >- Links from scrape run 1. Bright Data fires one run per snapshot, so for BD-produced rows this equals `appeared_links`. items: type: string appeared_links_run2: type: array description: >- Links from scrape run 2 (legacy Cloro double-run architecture). Empty for Bright Data rows. items: type: string my_citations: type: array nullable: true description: | URLs from the AI's answer that point at **your** domain (matched against the universe's `website`). items: type: string competitor_citations: type: array nullable: true description: | URLs from the AI's answer that point at **competitor** domains (matched against the universe's competitor list). items: type: string total_citations_count: type: integer description: >- Total cited URLs in this answer (yours + competitor + uncategorised). my_domain_citations_count: type: integer description: Count of citations to your domain. my_brand_mentions_count: type: integer description: >- Count of times the AI mentioned your brand by name in the answer text (regardless of citation). model_used: type: string nullable: true description: The model identifier the AI returned (e.g. `gpt-4o`, `gpt-5-3`). web_search_triggered: type: boolean nullable: true description: >- Whether the AI invoked web search while answering. Useful when comparing answers — search-backed answers tend to cite more. cost_milli_cents: type: integer nullable: true description: >- Cost of producing this row in 1/1000 of a cent (so 1234 = $0.01234). Internal accounting only — exposed so customers running their own cost tracking can reconcile. Error: type: object required: - error properties: error: type: object required: - type - message properties: type: type: string enum: - authentication_error - permission_error - not_found - invalid_request_error - rate_limit_error - api_error description: Machine-readable error category. message: type: string description: Human-readable explanation. responses: BadRequest: description: Malformed request (bad UUID, bad date, etc.). content: application/json: schema: $ref: '#/components/schemas/Error' Unauthorized: description: Missing / invalid / revoked API key. content: application/json: schema: $ref: '#/components/schemas/Error' Forbidden: description: API key is valid but doesn't own the requested universe. content: application/json: schema: $ref: '#/components/schemas/Error' NotFound: description: Universe (or metric, or row) doesn't exist. content: application/json: schema: $ref: '#/components/schemas/Error' securitySchemes: BearerAuth: type: http scheme: bearer bearerFormat: vso_<48 hex chars> description: | Generate a key at `/dashboard/settings/api-keys`. Pass it as `Authorization: Bearer vso_xxx` on every request. ````