> ## Documentation Index > Fetch the complete documentation index at: https://docs.hyperprop.com/llms.txt > Use this file to discover all available pages before exploring further. # Ask the indicator assistant > Sends the chat history and the current editor state to the indicator assistant and returns its answer. When the answer contains a script, `source` is the complete Hyperprop Script to put in the editor (declarations plus `compute`), with `pane` and `name` set when the assistant wants them changed. Pass `error` to have a failing script fixed. Requests are limited to 40 per trader per hour; the response is typically 5–30 seconds because the whole script is generated in one pass. Timeout after 120 seconds. ## OpenAPI ````yaml /api-reference/openapi.json post /v1/indicator-builder/chat openapi: 3.0.0 info: title: Hyperprop Platform API version: 1.0.0 description: >- REST API for the Hyperprop Trading Platform — provision evaluation and funded trading accounts, manage traders and plans, react to account lifecycle events via signed webhooks, and reconcile billing. Built for prop firms integrating from their own backend. Authentication, quick start, error handling, idempotency, pagination, custom metadata, webhooks, and the MCP connector are documented at https://docs.hyperprop.com. x-logo: url: https://app.hyperprop.com/logo-icon.svg altText: Hyperprop href: https://hyperprop.com servers: - url: https://api.hyperprop.com/platform description: Production security: [] tags: - name: Authentication description: >- User authentication - signup, signin, signout, password reset, email verification, OAuth, and MFA - name: User description: >- User account management - profile, notifications, agreements, dismissals, and audit logs. Requires Bearer token. - name: Organization description: >- Organization management - profile, team, trading accounts, plans, and rules. Supports **dual authentication**: Bearer JWT token (dashboard users) OR X-API-Key (programmatic access). - name: Market Data description: CME futures contracts and market data. Requires Bearer token. - name: Indicator Builder description: >- AI-assisted authoring of custom Hyperprop Script indicators for the web terminal. Requires Bearer token. - name: System description: System endpoints - health checks and connectivity paths: /v1/indicator-builder/chat: post: tags: - Indicator Builder summary: Ask the indicator assistant description: >- Sends the chat history and the current editor state to the indicator assistant and returns its answer. When the answer contains a script, `source` is the complete Hyperprop Script to put in the editor (declarations plus `compute`), with `pane` and `name` set when the assistant wants them changed. Pass `error` to have a failing script fixed. Requests are limited to 40 per trader per hour; the response is typically 5–30 seconds because the whole script is generated in one pass. Timeout after 120 seconds. operationId: postV1IndicatorbuilderChat requestBody: $ref: '#/components/requestBodies/Model470' responses: '200': description: Assistant reply content: '*/*': schema: $ref: '#/components/schemas/Model472' '400': description: Validation failed content: '*/*': schema: $ref: '#/components/schemas/Model473' '401': description: Authentication required content: '*/*': schema: $ref: '#/components/schemas/Model9' '429': description: >- You have reached the indicator builder limit for this hour. Try again later. content: '*/*': schema: $ref: '#/components/schemas/Model474' '502': description: The indicator assistant returned an error. content: '*/*': schema: $ref: '#/components/schemas/Model475' '503': description: The indicator assistant is not configured on this server. content: '*/*': schema: $ref: '#/components/schemas/Model476' '504': description: >- The indicator assistant took too long to answer. Try a simpler request. content: '*/*': schema: $ref: '#/components/schemas/Model477' components: requestBodies: Model470: content: application/json: schema: $ref: '#/components/schemas/Model470' schemas: Model472: type: object properties: success: type: boolean example: true data: $ref: '#/components/schemas/Model471' Model473: type: object properties: success: type: boolean description: Always false on errors example: false statusCode: type: number example: 400 error: type: string example: Bad Request message: type: string example: Validation failed code: type: string description: Machine-readable error code — switch on this, not on message text example: VALIDATION_ERROR Model9: type: object properties: success: type: boolean description: Always false on errors example: false statusCode: type: number example: 401 error: type: string example: Unauthorized message: type: string example: Authentication required code: type: string description: Machine-readable error code — switch on this, not on message text example: UNAUTHORIZED Model474: type: object properties: success: type: boolean description: Always false on errors example: false statusCode: type: number example: 429 error: type: string example: Too Many Requests message: type: string example: >- You have reached the indicator builder limit for this hour. Try again later. code: type: string description: Machine-readable error code — switch on this, not on message text example: RATE_LIMITED Model475: type: object properties: success: type: boolean description: Always false on errors example: false statusCode: type: number example: 502 error: type: string example: Error message: type: string example: The indicator assistant returned an error. code: type: string description: Machine-readable error code — switch on this, not on message text example: MODEL_ERROR Model476: type: object properties: success: type: boolean description: Always false on errors example: false statusCode: type: number example: 503 error: type: string example: Error message: type: string example: The indicator assistant is not configured on this server. code: type: string description: Machine-readable error code — switch on this, not on message text example: MODEL_UNAVAILABLE Model477: type: object properties: success: type: boolean description: Always false on errors example: false statusCode: type: number example: 504 error: type: string example: Error message: type: string example: >- The indicator assistant took too long to answer. Try a simpler request. code: type: string description: Machine-readable error code — switch on this, not on message text example: MODEL_TIMEOUT Model470: type: object properties: messages: $ref: '#/components/schemas/messages' script: $ref: '#/components/schemas/script' error: type: string description: >- Compile or runtime error the chart reported for the current script, if any. example: null maxLength: 2000 dataset: $ref: '#/components/schemas/dataset' required: - messages - script Model471: type: object properties: message: type: string description: Assistant prose for the chat panel. The code block has been removed. example: A 14-period RSI in its own pane with 30/70 guides. source: type: string description: Complete replacement script, or null when the answer needed no code. example: |- const length = input.int('Length', 14, { min: 2, max: 100 }); function compute(bars, inputs, hp) { const rsi = hp.rsi(bars.map(b => b.c), length); return { plots: [{ title: 'RSI', values: rsi, color: '#ab47bc' }], range: { min: 0, max: 100 }, guides: [30, 70] }; } pane: $ref: '#/components/schemas/pane' name: type: string example: RSI messages: type: array description: >- Chat history, oldest first. The last message must be from the user. Each message is capped at 24,000 characters — enough for a full Pine Script paste. example: - role: user content: Build an RSI with 30/70 guides. minItems: 1 maxItems: 40 items: $ref: '#/components/schemas/Model469' script: type: object description: >- What is currently in the editor. The reply replaces it wholesale when it contains code. properties: name: type: string example: RSI default: '' maxLength: 80 pane: $ref: '#/components/schemas/pane' source: type: string example: |- const length = input.int('Length', 14, { min: 2, max: 100 }); function compute(bars, inputs, hp) { const rsi = hp.rsi(bars.map(b => b.c), length); return { plots: [{ title: 'RSI', values: rsi, color: '#ab47bc' }], range: { min: 0, max: 100 }, guides: [30, 70] }; } maxLength: 40000 required: - pane - source dataset: type: object description: >- The sample series the builder chart is showing, so the model can reason about scale and bar count. properties: symbol: type: string example: ES maxLength: 16 interval: type: string example: 5m maxLength: 8 barCount: type: integer example: 3000 minimum: 0 maximum: 100000 required: - symbol - interval - barCount pane: type: string example: separate enum: - overlay - separate Model469: type: object properties: role: $ref: '#/components/schemas/Model468' content: type: string minLength: 1 maxLength: 24000 required: - role - content Model468: type: string enum: - user - assistant ```` ## Related topics - [Ask the indicator assistant (streaming)](/platform-api/indicator-builder/ask-the-indicator-assistant-streaming.md) - [Indicator assistant availability](/platform-api/indicator-builder/indicator-assistant-availability.md) - [Hyperprop Script: build indicators (agent recipe)](/guides/hyperprop-script.md) - [Changelog](/changelog.md) - [Get all order books (best bid/ask)](/market-data-api/order-book/get-all-order-books-best-bidask.md)