> ## Documentation Index > Fetch the complete documentation index at: https://help.onlyx.ai/llms.txt > Use this file to discover all available pages before exploring further. > ## Agent Instructions > You are reading OnlyX Help: the OnlyX Help Center and the OnlyX developer documentation. OnlyX is an AI chatting and CRM platform for OnlyFans agencies. Its AI chatter is called Hugo in the app. > Pages at the site root (for example /inbox/..., /hugo/..., /billing/...) are Help Center articles for agency owners, admins, chatters and creators who use the app at app.onlyx.ai. Words in bold are the exact button, menu and label names the app shows; keep them exactly as written. When these pages do not answer a question, the person can email support at support@onlyx.ai. > Pages under /developers are the developer documentation. REST API base URL: https://api.onlyx.ai/v1 (authenticate with `Authorization: Bearer `; keys start with onx_sk_ and are created in app.onlyx.ai under Settings > API & MCP). MCP server: https://mcp.onlyx.ai/mcp (OAuth, or a Bearer API key). In the API the AI chatter is `ai` on the wire. Money is integer US cents in fields ending in Cents; timestamps are UTC ISO-8601. > Rules for assistants acting on a user's behalf: discover ids with list calls and never invent them; before any call that reaches a real fan or the live OnlyFans account (sending a message, releasing a chat to the AI, turning AI on for a chat, resolving a hand-off, turning review mode off, changing the welcome message, creating a tracking link) show the user the exact content and get explicit confirmation; send every POST with an Idempotency-Key and reuse it on retry; never resend a message whose delivery status is unconfirmed; never ask a creator for her OnlyFans password or codes - she signs in herself through a connect link and the OnlyX Login app. # Update AI settings > Changes the AI switch, follow-ups or review mode; send only what changes. These settings reach real fans: `aiEnabled: true` lets the AI answer every waiting fan, follow-ups are real messages, and **turning review mode off releases every waiting draft to fans immediately** — except drafts that have waited longer than 10 minutes, which are discarded rather than sent late. Because of that, turning `review.enabled` from on to off also needs the `messages:send` scope (`403 INSUFFICIENT_SCOPE` otherwise) and is limited to 6 per hour per creator. Confirm with a person before changing any of them on someone's behalf. **Scope:** requires `ai:write`; turning `review.enabled` from on to off also requires `messages:send`. **Rate limits:** 6 review-off switches per hour per creator, across all credentials — on top of the general limit of 120 requests per 60 seconds per credential. **Idempotency:** send an `Idempotency-Key` header to make retries safe: a replay within 24 hours returns the original response with `Idempotent-Replayed: true`. **Reaches fans:** this operation can reach a real fan or change the live OnlyFans account. Confirm it with a person before calling it on their behalf. ## OpenAPI ````yaml /developers/api-reference/openapi.json patch /v1/creators/{creatorId}/ai-settings openapi: 3.1.0 info: contact: email: developers@onlyx.ai name: OnlyX developers url: https://docs.onlyx.ai/ description: > The OnlyX API runs your OnlyX workspace from your own code and AI tools: creators and their connection, the fan inbox, fans, vault media, statistics, the AI chatter's setup and tracking links. **Authentication.** Send an API key as `Authorization: Bearer onx_sk_…` (create one in OnlyX under Settings → API & MCP), or an OAuth access token issued to a connected app. Keys belong to the workspace, carry scopes, and can be limited to some creators. **Conventions.** JSON with camelCase fields; timestamps are ISO-8601 UTC with a `Z`; money is integer US cents (fields end in `Cents`). Lists return `{"data": [...], "hasMore": bool, "nextCursor": string|null}` and page with `limit` and `cursor`. **Scopes.** Every operation names the scope it needs in `x-required-scope` and in its description; a credential without it gets `403 INSUFFICIENT_SCOPE`. **Errors and limits.** Errors are `{"error": {"code", "message", "requestId"}}`; every response carries `X-Request-Id`. The default limit is 120 requests per 60 seconds per credential, reported in `X-RateLimit-*` headers; over it you get `429 RATE_LIMITED` with `Retry-After`. Operations with limits of their own list them in `x-rate-limits`. **Idempotency and safety.** Every write accepts an `Idempotency-Key` header (required when sending a message, adding a creator and creating a tracking link): retry with the same key and you get the original answer instead of a second action. Operations marked `x-reaches-fans: true` can reach a real fan or change the live OnlyFans account. Guides, the MCP server and more: [docs.onlyx.ai](https://docs.onlyx.ai). summary: >- Run your OnlyX workspace (creators, inbox, fans, stats and the AI chatter) from code and AI tools. title: OnlyX API version: 1.0.0 servers: - description: Production url: https://api.onlyx.ai security: - bearerAuth: [] tags: - description: 'Who you are: the workspace and credential behind a key.' name: Workspace - description: >- The creators your workspace manages: list, read, add, rename, and the AI switch. name: Creators - description: >- Connecting a creator's OnlyFans account: connect links she opens on her own device, and the connection status to poll. name: Connect - description: >- The fan inbox: list and count conversations, read state, take over, release, and the per-chat AI switch. name: Conversations - description: Read message history and send text, free media and paid messages to fans. name: Messages - description: Conversations the AI chatter handed to your team, and resolving them. name: Hand-offs - description: >- Fans, their purchases, fan lists, and your team's notes, custom names and mute flag. name: Fans - description: >- Each creator's OnlyFans vault: media ids for messages and levels, and thumbnails. name: Media - description: >- Today, revenue, audience, the cached overview report, and each creator's transaction ledger. name: Stats - description: The brief the AI chatter follows to chat as each creator. name: AI persona - description: 'The AI content ladder: folders, ladders, priced levels and their media.' name: AI content - description: >- Follow-ups, review mode, the master AI switch, and OnlyFans' welcome message. name: AI settings - description: >- OnlyFans tracking and trial links, your cost fields, and creating new links. name: Tracking links externalDocs: description: OnlyX developer documentation url: https://docs.onlyx.ai paths: /v1/creators/{creatorId}/ai-settings: patch: tags: - AI settings summary: Update AI settings description: >- Changes the AI switch, follow-ups or review mode; send only what changes. These settings reach real fans: `aiEnabled: true` lets the AI answer every waiting fan, follow-ups are real messages, and **turning review mode off releases every waiting draft to fans immediately** — except drafts that have waited longer than 10 minutes, which are discarded rather than sent late. Because of that, turning `review.enabled` from on to off also needs the `messages:send` scope (`403 INSUFFICIENT_SCOPE` otherwise) and is limited to 6 per hour per creator. Confirm with a person before changing any of them on someone's behalf. **Scope:** requires `ai:write`; turning `review.enabled` from on to off also requires `messages:send`. **Rate limits:** 6 review-off switches per hour per creator, across all credentials — on top of the general limit of 120 requests per 60 seconds per credential. **Idempotency:** send an `Idempotency-Key` header to make retries safe: a replay within 24 hours returns the original response with `Idempotent-Replayed: true`. **Reaches fans:** this operation can reach a real fan or change the live OnlyFans account. Confirm it with a person before calling it on their behalf. operationId: updateAiSettings parameters: - description: The creator id (`cre_…`). in: path name: creatorId required: true schema: description: The creator id (`cre_…`). title: Creatorid type: string - $ref: '#/components/parameters/IdempotencyKey' requestBody: content: application/json: schema: $ref: '#/components/schemas/UpdateAiSettingsRequest' required: true responses: '200': content: application/json: example: aiEnabled: true followUps: effective: true enabled: true maxPerFanPerDay: 2 plannedToday: 14 sentToday: 9 unansweredAfterHours: 24 massMessages: effective: true enabled: true nextSendAt: '2026-09-26T18:00:00.000Z' perDay: 1 review: enabled: false reviewMinutes: 5 waitingNow: 0 whenUnreviewed: send schema: $ref: '#/components/schemas/AiSettings' description: The AI settings after the change. headers: Idempotent-Replayed: $ref: '#/components/headers/Idempotent-Replayed' X-RateLimit-Limit: $ref: '#/components/headers/X-RateLimit-Limit' X-RateLimit-Remaining: $ref: '#/components/headers/X-RateLimit-Remaining' X-RateLimit-Reset: $ref: '#/components/headers/X-RateLimit-Reset' X-Request-Id: $ref: '#/components/headers/X-Request-Id' '400': content: application/json: examples: VALIDATION_ERROR: summary: VALIDATION_ERROR value: error: code: VALIDATION_ERROR message: 'aiEnabled: Input should be a valid boolean' requestId: req-3f9a1c2b7d4e5f60a1b2c3d4 schema: $ref: '#/components/schemas/Error' description: >- The request is not valid and was not carried out. Fix it before retrying. `VALIDATION_ERROR`: the request is not valid; `message` names the field and the problem. headers: X-Request-Id: $ref: '#/components/headers/X-Request-Id' '401': $ref: '#/components/responses/Unauthorized' '403': content: application/json: examples: INSUFFICIENT_SCOPE: summary: INSUFFICIENT_SCOPE value: error: code: INSUFFICIENT_SCOPE message: This key does not have the `ai:write` scope. requestId: req-3f9a1c2b7d4e5f60a1b2c3d4 INSUFFICIENT_SCOPE_MESSAGES_SEND: summary: INSUFFICIENT_SCOPE (messages:send) value: error: code: INSUFFICIENT_SCOPE message: This key does not have the `messages:send` scope. requestId: req-3f9a1c2b7d4e5f60a1b2c3d4 WORKSPACE_SUSPENDED: summary: WORKSPACE_SUSPENDED value: error: code: WORKSPACE_SUSPENDED message: This workspace is suspended. requestId: req-3f9a1c2b7d4e5f60a1b2c3d4 schema: $ref: '#/components/schemas/Error' description: >- The credential may not do this. `INSUFFICIENT_SCOPE`: the credential lacks the `ai:write` scope (the `WWW-Authenticate` header names it; an OAuth token's message says "token" instead of "key"). `INSUFFICIENT_SCOPE`: turning `review.enabled` from on to off without the `messages:send` scope. `WORKSPACE_SUSPENDED`: the workspace is suspended. headers: WWW-Authenticate: $ref: '#/components/headers/WWW-Authenticate' X-Request-Id: $ref: '#/components/headers/X-Request-Id' '404': content: application/json: examples: CREATOR_NOT_FOUND: summary: CREATOR_NOT_FOUND value: error: code: CREATOR_NOT_FOUND message: Creator not found. requestId: req-3f9a1c2b7d4e5f60a1b2c3d4 schema: $ref: '#/components/schemas/Error' description: >- Not found. An id that does not exist, belongs to another workspace, or is outside this credential's creators all get the same answer. `CREATOR_NOT_FOUND`: no creator with this `creatorId` is visible to this credential. headers: X-Request-Id: $ref: '#/components/headers/X-Request-Id' '409': content: application/json: examples: IDEMPOTENCY_IN_PROGRESS: summary: IDEMPOTENCY_IN_PROGRESS value: error: code: IDEMPOTENCY_IN_PROGRESS message: >- A request with this Idempotency-Key is still being processed. Retry in a few seconds. requestId: req-3f9a1c2b7d4e5f60a1b2c3d4 IDEMPOTENCY_KEY_REUSED: summary: IDEMPOTENCY_KEY_REUSED value: error: code: IDEMPOTENCY_KEY_REUSED message: >- This Idempotency-Key was already used for a different request. Use a new key for a new request. requestId: req-3f9a1c2b7d4e5f60a1b2c3d4 schema: $ref: '#/components/schemas/Error' description: >- The request conflicts with the current state. `IDEMPOTENCY_KEY_REUSED`: this `Idempotency-Key` was already used for a different request (or by another credential). Use a new key for a new request. `IDEMPOTENCY_IN_PROGRESS`: a request with this `Idempotency-Key` is still running. Retry in a few seconds with the same key. headers: X-Request-Id: $ref: '#/components/headers/X-Request-Id' '429': content: application/json: examples: RATE_LIMITED: summary: RATE_LIMITED (general limit) value: error: code: RATE_LIMITED message: Too many requests. Try again in 12 seconds. requestId: req-3f9a1c2b7d4e5f60a1b2c3d4 RATE_LIMITED_1: summary: >- RATE_LIMITED (6 review-off switches per hour per creator, across all credentials) value: error: code: RATE_LIMITED message: >- Review mode can be turned off at most 6 times an hour for a creator, because each time releases every waiting draft to fans. Try again later. requestId: req-3f9a1c2b7d4e5f60a1b2c3d4 schema: $ref: '#/components/schemas/Error' description: >- Over a rate limit (`RATE_LIMITED`): the general limit, or this operation's own (6 review-off switches per hour per creator, across all credentials). Wait `Retry-After` seconds, then retry — with the same `Idempotency-Key` for a write. headers: Retry-After: $ref: '#/components/headers/Retry-After' X-RateLimit-Limit: $ref: '#/components/headers/X-RateLimit-Limit' X-RateLimit-Remaining: $ref: '#/components/headers/X-RateLimit-Remaining' X-RateLimit-Reset: $ref: '#/components/headers/X-RateLimit-Reset' X-Request-Id: $ref: '#/components/headers/X-Request-Id' '500': $ref: '#/components/responses/InternalError' components: parameters: IdempotencyKey: description: >- Optional. A unique value (8 to 64 letters, digits, `-` or `_`; a UUID is ideal) that makes a retry safe: the same key with the same request within 24 hours returns the stored response, with `Idempotent-Replayed: true`, instead of doing the work twice. The same key with a different request is `409 IDEMPOTENCY_KEY_REUSED`; while the first request still runs it is `409 IDEMPOTENCY_IN_PROGRESS`. A malformed key is `400 VALIDATION_ERROR`. example: 3b1f7c2e-9d4a-4e8b-a6c1-5f0d2e7b9a14 in: header name: Idempotency-Key required: false schema: maxLength: 64 minLength: 8 pattern: ^[A-Za-z0-9_-]{8,64}$ type: string schemas: UpdateAiSettingsRequest: additionalProperties: false description: >- Send only what you want to change. Mass messages cannot be changed through the API in v1. examples: - followUps: maxPerFanPerDay: 1 - review: enabled: true reviewMinutes: 10 whenUnreviewed: hold properties: aiEnabled: description: >- The creator's master AI switch. Turning it on lets the AI answer every waiting fan in her AI chats. type: boolean followUps: $ref: '#/components/schemas/FollowUpsPatch' review: $ref: '#/components/schemas/ReviewPatch' title: UpdateAiSettingsRequest type: object AiSettings: description: >- How the AI chatter behaves for one creator. New fields may be added to this report; ignore fields you do not know. examples: - aiEnabled: true followUps: effective: true enabled: true maxPerFanPerDay: 2 plannedToday: 14 sentToday: 9 unansweredAfterHours: 24 massMessages: effective: true enabled: true nextSendAt: '2026-09-26T18:00:00.000Z' perDay: 1 review: enabled: false reviewMinutes: 5 waitingNow: 0 whenUnreviewed: send properties: aiEnabled: description: >- The creator's master AI switch (the same value as `aiEnabled` on the Creator). type: boolean followUps: $ref: '#/components/schemas/FollowUps' massMessages: $ref: '#/components/schemas/MassMessages' review: $ref: '#/components/schemas/Review' required: - aiEnabled - followUps - review - massMessages title: AiSettings type: object Error: properties: error: properties: code: description: Stable, machine-readable error code. type: string message: description: What went wrong, in a sentence for a person. type: string requestId: description: >- The id of this request (also in the `X-Request-Id` header). Quote it to support. type: string required: - code - message - requestId type: object required: - error type: object FollowUpsPatch: additionalProperties: false properties: enabled: description: Switch follow-ups on or off. type: boolean maxPerFanPerDay: description: 1 to 4. maximum: 4 minimum: 1 type: integer title: FollowUpsPatch type: object ReviewPatch: additionalProperties: false properties: enabled: description: >- Switch review mode. **Turning it off releases every waiting draft to fans immediately** (drafts that have waited longer than 10 minutes are discarded instead of sent late), so it also needs the `messages:send` scope and is limited to 6 per hour per creator. type: boolean reviewMinutes: description: 1 to 60. maximum: 60 minimum: 1 type: integer whenUnreviewed: description: '`send` or `hold`.' enum: - send - hold type: string title: ReviewPatch type: object FollowUps: description: >- The AI's own follow-up messages to fans who went quiet. These are real messages to fans. properties: effective: description: >- Whether follow-ups actually go out now: `enabled` and the creator's AI is on. type: boolean enabled: description: Your follow-up setting. type: boolean maxPerFanPerDay: description: At most this many follow-ups per fan per day, 1 to 4. type: integer plannedToday: description: Follow-ups planned today (her time zone). type: integer sentToday: description: Follow-ups sent today (her time zone). type: integer unansweredAfterHours: description: How long a fan stays silent before a follow-up is considered. type: integer required: - enabled - effective - maxPerFanPerDay - unansweredAfterHours - plannedToday - sentToday title: FollowUps type: object MassMessages: description: >- The AI's scheduled messages to fan lists. Read-only in v1: configure them in the dashboard. properties: effective: description: Whether mass messages actually go out now. type: boolean enabled: description: Your mass-message setting. type: boolean nextSendAt: anyOf: - type: string - type: 'null' description: When the next one is planned, or `null`. perDay: description: Mass messages per day. type: integer required: - enabled - effective - perDay - nextSendAt title: MassMessages type: object Review: description: 'Review mode: whether every AI reply waits for a person to approve it.' properties: enabled: description: '`true` makes every AI reply wait for a person to review it.' type: boolean reviewMinutes: description: How long a reply waits for review, 1 to 60 minutes. type: integer waitingNow: description: AI replies waiting for review right now. type: integer whenUnreviewed: description: 'What happens when nobody reviews in time: `send` it, or `hold` it.' enum: - send - hold type: string required: - enabled - reviewMinutes - whenUnreviewed - waitingNow title: Review type: object headers: Idempotent-Replayed: description: >- `true` when this response is the stored answer to an earlier request with the same `Idempotency-Key`: nothing was done again. Absent on a first answer. schema: enum: - 'true' type: string X-RateLimit-Limit: description: >- Requests allowed in the current window of the tightest limit this request counted against. example: 120 schema: type: integer X-RateLimit-Remaining: description: Requests left in that window. example: 117 schema: type: integer X-RateLimit-Reset: description: When that window resets, as Unix epoch seconds. example: 1790431380 schema: type: integer X-Request-Id: description: >- The id of this request. Your own `X-Request-Id` (8 to 64 letters, digits or `-`) is echoed back; otherwise OnlyX creates one. Also in every error body as `requestId`: quote it to support. example: req-3f9a1c2b7d4e5f60a1b2c3d4 schema: type: string WWW-Authenticate: description: >- The authentication challenge (RFC 6750), for example `Bearer error="insufficient_scope", scope="messages:send"`. schema: type: string Retry-After: description: Seconds to wait before trying again. example: 12 schema: type: integer responses: Unauthorized: content: application/json: examples: INVALID_API_KEY: summary: INVALID_API_KEY value: error: code: INVALID_API_KEY message: The API key is invalid, expired or revoked. requestId: req-3f9a1c2b7d4e5f60a1b2c3d4 INVALID_TOKEN: summary: INVALID_TOKEN value: error: code: INVALID_TOKEN message: The access token is invalid, expired or revoked. requestId: req-3f9a1c2b7d4e5f60a1b2c3d4 UNAUTHORIZED: summary: UNAUTHORIZED value: error: code: UNAUTHORIZED message: >- Authentication required. Send your API key as `Authorization: Bearer onx_sk_…`. requestId: req-3f9a1c2b7d4e5f60a1b2c3d4 schema: $ref: '#/components/schemas/Error' description: >- No credential was sent (`UNAUTHORIZED`), or it is invalid, expired or revoked (`INVALID_API_KEY` for API keys, `INVALID_TOKEN` for OAuth access tokens: refresh the token or reconnect the app). Too many failed attempts from one address with credentials OnlyX never issued are answered `429 RATE_LIMITED` instead. headers: WWW-Authenticate: $ref: '#/components/headers/WWW-Authenticate' X-Request-Id: $ref: '#/components/headers/X-Request-Id' InternalError: content: application/json: examples: INTERNAL_ERROR: summary: INTERNAL_ERROR value: error: code: INTERNAL_ERROR message: Something went wrong on our side. requestId: req-3f9a1c2b7d4e5f60a1b2c3d4 schema: $ref: '#/components/schemas/Error' description: >- Something went wrong on OnlyX's side (`INTERNAL_ERROR`). Retry with backoff (reuse the `Idempotency-Key` for a write); quote `requestId` to support if it persists. headers: X-Request-Id: $ref: '#/components/headers/X-Request-Id' securitySchemes: bearerAuth: bearerFormat: onx_sk_… API key or onx_at_… access token description: >- An API key (`onx_sk_…`) created in OnlyX under Settings → API & MCP, or an OAuth access token (`onx_at_…`) issued to a connected app. Send it as `Authorization: Bearer `. API keys may also be sent as `X-API-Key: `. scheme: bearer type: http ````