> For the complete documentation index, see [llms.txt](https://docs.umbraco.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.umbraco.com/ai-in-umbraco/19.latest/management-api/decision/ask-decision.md). # Ask Decision Ask a Decision profile a yes/no, pick-one, or score question and get a typed answer. {% hint style="warning" %} Returns **404** with an empty body when the `Umbraco:AI:Experimental:Decision` feature flag is disabled. The flag is checked before the request body is read or validated. {% endhint %} ```http POST /umbraco/ai/management/api/v1/decision/ask ``` ## Request Body The `state` property holds the content being judged. The `question` object is polymorphic. Its `$type` property selects the kind of question: `binary`, `choice`, or `score`. ### Binary Question {% code title="application/json" %} ```json { "profileIdOrAlias": "moderation", "state": "Buy cheap watches at example.com", "question": { "$type": "binary", "instructions": "Is this comment spam?", "trueCriteria": "Promotional, a scam, or unrelated links", "falseCriteria": "A genuine comment from a reader" } } ``` {% endcode %} ### Choice Question {% code title="application/json" %} ```json { "state": "I was charged twice for my subscription.", "question": { "$type": "choice", "instructions": "Which team should handle this message?", "options": [ { "key": "billing", "description": "Invoices, payments, and refunds" }, { "key": "technical", "description": "Errors, bugs, and how-to questions" }, { "key": "sales" } ] } } ``` {% endcode %} ### Score Question {% code title="application/json" %} ```json { "state": "Our new plans give you more storage for the same price.", "question": { "$type": "score", "instructions": "How easy is this summary to read for a general audience?", "levels": [ { "description": "poor" }, { "description": "ok" }, { "description": "good" } ] } } ``` {% endcode %} | Property | Type | Required | Description | | ------------------------ | ------ | ----------- | ------------------------------------------------------------------------------------------------------------------------------ | | `profileIdOrAlias` | string | No | Profile ID or alias. Uses the default Decision profile if omitted. | | `state` | string | No | The content being judged, such as a comment or a page body. | | `question` | object | Yes | The question to ask. | | `question.$type` | string | Yes | `"binary"`, `"choice"`, or `"score"`. Must be the first property of `question`. | | `question.instructions` | string | Yes | What to decide. Must not be empty. | | `question.trueCriteria` | string | No | Binary only. What counts as "yes". | | `question.falseCriteria` | string | No | Binary only. What counts as "no". | | `question.options` | array | Choice only | 2 to 255 options. Each has a unique, non-empty `key` and an optional `description`. Keys are compared exactly, including case. | | `question.levels` | array | Score only | 2 to 10 levels, lowest first. Each has a non-empty `description`. | ## Response The response `$type` matches the question's `$type`. `confidence` is left out when the provider does not report one. ### Binary Response {% code title="200 OK" %} ```json { "$type": "binary", "trueProbability": 0.97, "modelId": "jev-latest", "usage": { "inputTokens": 42, "outputTokens": 1, "totalTokens": 43 } } ``` {% endcode %} ### Choice Response {% code title="200 OK" %} ```json { "$type": "choice", "choice": "billing", "confidence": 0.91, "probabilities": { "billing": 0.91, "technical": 0.07, "sales": 0.02 }, "modelId": "jev-latest", "usage": { "inputTokens": 58, "outputTokens": 1, "totalTokens": 59 } } ``` {% endcode %} ### Score Response {% code title="200 OK" %} ```json { "$type": "score", "score": 1.8, "confidence": 0.8, "probabilities": { "0": 0.05, "1": 0.15, "2": 0.8 }, "modelId": "jev-latest", "usage": { "inputTokens": 51, "outputTokens": 1, "totalTokens": 52 } } ``` {% endcode %} | Property | Type | Applies to | Description | | ----------------- | ------ | ------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | `$type` | string | All | `"binary"`, `"choice"`, or `"score"`. | | `trueProbability` | number | Binary | The estimated probability that the answer is "yes", from 0.0 to 1.0. Apply your own threshold to turn it into a yes/no answer. | | `choice` | string | Choice | The `key` of the chosen option. | | `score` | number | Score | A 0-based, fractional index into `levels`. | | `confidence` | number | Choice, Score | The provider's own confidence in the answer, from 0.0 to 1.0. Left out when the provider does not report one. | | `probabilities` | object | Choice, Score | The estimated probability for every option or level. Choice probabilities are keyed by option `key`. Score probabilities are keyed by level index, from `"0"` for the lowest level. | | `modelId` | string | All | The model that produced the answer. `null` when unknown. | | `usage` | object | All | Token usage (`inputTokens`, `outputTokens`, `totalTokens`). `null` when the provider does not report usage. | ## Status Codes | Code | Meaning | | ---- | --------------------------------------------------------------------------------------------------------------- | | 200 | The question was answered. | | 400 | The question is invalid, or no answer could be produced. See below. | | 404 | The feature flag is disabled (empty body), or `profileIdOrAlias` did not match a profile (`Profile not found`). | A `400` response returns `ProblemDetails` in these cases: * The request body is missing required properties, or `$type` is missing, unknown, or not the first property of `question`. * `instructions` is empty. * A choice question has fewer than 2 or more than 255 options, or an option key is empty or duplicated. * A score question has fewer than 2 or more than 10 levels, or a level is empty. * The profile is not a Decision profile. * `profileIdOrAlias` is omitted and no default Decision profile is configured. * The provider call failed, for example because of an invalid API key, rate limiting, or a request the provider rejected. * The provider's answer was incomplete or inconsistent, for example a missing probability or a choice that is not one of the keys. The call is recorded as failed. Invalid questions are rejected before any provider is called. ## Example {% code title="cURL" %} ```bash curl -X POST "https://your-site.com/umbraco/ai/management/api/v1/decision/ask" \ -H "Authorization: Bearer YOUR_ACCESS_TOKEN" \ -H "Content-Type: application/json" \ -d '{ "state": "Buy cheap watches", "question": { "$type": "binary", "instructions": "Is this comment spam?" } }' ``` {% endcode %} ## Related * [Decision API overview](/ai-in-umbraco/19.latest/management-api/decision.md) * [Using the Decision API](/ai-in-umbraco/19.latest/using-the-api/decision.md) * [Decision Controller](/ai-in-umbraco/19.latest/frontend/decision-controller.md) --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation by asking a question. Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters: ``` GET https://docs.umbraco.com/ai-in-umbraco/19.latest/management-api/decision/ask-decision.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.