> ## Documentation Index > Fetch the complete documentation index at: https://quantura.mintlify.site/llms.txt > Use this file to discover all available pages before exploring further. # Ask Scout > Scout classifies topic, quantile and statistic. Quantura computes facts from immutable input/forecast values, up to 500 history rows and 512 forecast steps. It does not generate trading returns or infer win rates. Reuse conversation_id and turn_id on retries; a completed turn returns its saved response without another model call. Each conversation has up to 30 turns. Web sessions require sign-in; programmatic calls require active metered or legacy Pro access (including an active 14-day trial), enterprise access or the verified admin. Owner/resource access is rechecked on every call. ## OpenAPI ````yaml /docs/api/openapi.json post /v1/jev/forecast-questions openapi: 3.1.0 info: title: Quantura API version: 1.0.0 description: >- Search markets and public datasets, download historical observations, create forecasts and inspect screener quantiles. contact: name: Quantura url: https://quantura.studio/contact license: name: Quantura API Terms url: https://quantura.studio/terms servers: - url: https://quantura.studio/api security: - bearerAuth: [] - clerkOAuth: - openid - profile - email tags: - name: Access - name: Forecast - name: Screener - name: Search - name: Download - name: Scout paths: /v1/jev/forecast-questions: servers: - url: https://quantura.studio/api post: tags: - Scout summary: Ask Scout description: >- Scout classifies topic, quantile and statistic. Quantura computes facts from immutable input/forecast values, up to 500 history rows and 512 forecast steps. It does not generate trading returns or infer win rates. Reuse conversation_id and turn_id on retries; a completed turn returns its saved response without another model call. Each conversation has up to 30 turns. Web sessions require sign-in; programmatic calls require active metered or legacy Pro access (including an active 14-day trial), enterprise access or the verified admin. Owner/resource access is rechecked on every call. operationId: askJevForecastQuestion requestBody: required: true content: application/json: schema: type: object additionalProperties: false required: - context - question - turn_id properties: context: $ref: '#/components/schemas/JevForecastContext' question: type: string minLength: 1 maxLength: 1500 conversation_id: type: string format: uuid turn_id: type: string format: uuid example: context: kind: ensemble id: ens_saved_forecast question: What is the maximum P50 across this forecast? conversation_id: 5c71cfa6-57f8-4a69-941d-73cadfd6d263 turn_id: e5eb1e65-881d-412a-a31b-5cf1b72a3eaf responses: '200': description: Saved structured answer with computed facts and source metadata. content: application/json: schema: type: object example: data: conversation_id: 5c71cfa6-57f8-4a69-941d-73cadfd6d263 question: What is the maximum P50 across this forecast? response: schema_version: forecast_question_v1 model: jev-1.13.0 response_mode: typed_routing_computed_facts heading: Forecast values topic: values answer: Values come from the saved forecast. facts: - label: max P50 value: 346.81 kind: forecast timestamp: '2026-10-06T00:00:00Z' '401': description: Sign-in required or invalid authentication. '403': description: Paid API entitlement, scope or current resource access denied. '404': description: Saved or published context unavailable. '409': description: Context changed, conversation busy, or 30-turn limit reached. '422': description: Invalid reference, question, turn ID or annotation. '429': description: Per-user or global Scout budget reached. '503': description: Scout unavailable; no substitute answer is generated. security: - bearerAuth: [] - clerkOAuth: - openid - profile - email components: schemas: JevForecastContext: oneOf: - type: object additionalProperties: false required: - kind - id properties: kind: type: string enum: - ensemble - game - saved_game - sagemaker id: type: string pattern: ^[A-Za-z0-9_-]{1,220}$ - type: object additionalProperties: false required: - kind - symbol - scan_id properties: kind: const: screener symbol: type: string maxLength: 40 scan_id: type: string maxLength: 160 - type: object additionalProperties: false required: - kind - name - frequency - rows properties: kind: const: preview name: type: string minLength: 1 maxLength: 120 frequency: type: string maxLength: 40 rows: type: array minItems: 2 maxItems: 500 items: type: object required: - timestamp - target properties: timestamp: type: string format: date-time target: type: number description: >- References are resolved and authorized server-side. Client-supplied forecast arrays are rejected. Preview rows are explicitly user-supplied observations, not verified provider data. securitySchemes: bearerAuth: type: http scheme: bearer bearerFormat: Clerk personal API key or legacy Quantura key clerkOAuth: type: oauth2 description: >- Clerk authorization code with PKCE S256. Pass resource=https://quantura.studio on authorization and token requests. Paid Pro, enterprise or verified administrator API access is required. flows: authorizationCode: authorizationUrl: https://clerk.quantura.studio/oauth/authorize tokenUrl: https://clerk.quantura.studio/oauth/token refreshUrl: https://clerk.quantura.studio/oauth/token scopes: openid: Identify the Quantura account profile: Resolve the account profile email: Verify account email offline_access: Refresh access after expiry ```` This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.