> ## Documentation Index
> Fetch the complete documentation index at: https://docs.sublay.io/llms.txt
> Use this file to discover all available pages before exploring further.

# Ask content

> Ask a natural language question and stream a generated answer from project content

## Overview

Streams an AI-generated answer to a natural language question, grounded in your project's content. Uses SSE (Server-Sent Events) to deliver tokens incrementally. Returns source citations once the stream completes.

For integration guidance, see [Ask Content](/sdk/search/ask).

## Usage Example

```tsx theme={null}
import { useAskContent } from "@sublay/react-js";

function AskUI() {
  const { answer, sources, streaming, loading, error, ask, reset } = useAskContent();

  return (
    <div>
      <button onClick={() => ask({ query: "How do I reset my password?" })} disabled={loading || streaming}>
        Ask
      </button>
      {answer && <p>{answer}{streaming && " ▋"}</p>}
      {sources.map((s) => <div key={s.record.id}>[{s.sourceType}] {s.record.id}</div>)}
    </div>
  );
}
```

## Parameters

Call `ask` with:

<ParamField path="query" type="string" required>
  The natural language question.
</ParamField>

<ParamField path="sourceTypes" type="(&#x22;entity&#x22; | &#x22;comment&#x22; | &#x22;message&#x22;)[]">
  Limit the content types searched for context.
</ParamField>

<ParamField path="spaceId" type="string">
  Scope context lookup to a specific space.
</ParamField>

<ParamField path="includeChildSpaces" type="boolean" default="false">
  Only applies with `spaceId`. When `true`, the context lookup also covers every space nested under it — children, grandchildren, the whole subtree at any depth — so you can ask across a community and all of its channels at once. Ignored without `spaceId`.
</ParamField>

<ParamField path="conversationId" type="string">
  Scope context lookup to a specific conversation.
</ParamField>

<ParamField path="limit" type="number">
  Maximum number of source records to use as context.
</ParamField>

<SpaceReputationParams variant="context" field="path" />

## Returns

<ResponseField name="answer" type="string">
  The AI-generated answer. Grows token by token while `streaming` is `true`.
</ResponseField>

<ResponseField name="sources" type="ContentSearchResult[]">
  Source records used as context. Populated after streaming ends. When `spaceReputation` is requested, each record's embedded author carries an added `spaceReputation` number. See [Reputation](/data-models/reputation).
</ResponseField>

<ResponseField name="streaming" type="boolean">
  `true` while the SSE stream is open and tokens are arriving.
</ResponseField>

<ResponseField name="loading" type="boolean">
  `true` from `ask()` call until the first token arrives.
</ResponseField>

<ResponseField name="error" type="string | null">
  Error message if the request failed.
</ResponseField>

<ResponseField name="ask" type="(props) => void">
  Starts a new question. Aborts any in-flight stream first.
</ResponseField>

<ResponseField name="reset" type="() => void">
  Aborts in-flight stream and clears all state.
</ResponseField>

<Note>
  In React Native, SSE streaming requires polyfills for `fetch`, `ReadableStream`, and `TextDecoder`. Install `react-native-fetch-api`, `web-streams-polyfill@^3`, `react-native-polyfill-globals`, and `text-encoding`, then call the targeted sub-polyfills once at app startup, before using this hook:

  ```ts theme={null}
  import { polyfill as polyfillEncoding } from "react-native-polyfill-globals/src/encoding";
  import { polyfill as polyfillReadableStream } from "react-native-polyfill-globals/src/readable-stream";
  import { polyfill as polyfillFetch } from "react-native-polyfill-globals/src/fetch";

  polyfillEncoding();
  polyfillReadableStream();
  polyfillFetch(); // last, so it wins over the built-in fetch
  ```

  Prefer these targeted polyfills over the all-in-one `polyfillGlobals()`: that helper also pulls in `react-native-get-random-values`, a native module that forces a development-client rebuild. The three above are pure JavaScript, so a Metro reload is enough. Pin `web-streams-polyfill` to v3 — v4 removed the `ponyfill/es6` subpath that `react-native-polyfill-globals` requires.
</Note>