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

# Ask and Continue

> Ask any model in the Nito catalog from inside your current chat, then continue the exchange.

`ask` sends one prompt to a model through Nito and returns the answer inside your current chat. `continue` follows up on that same exchange. Both make **at most one** Nito call.

`ask` is a normal turn in your harness chat, so **your harness model can see the ask prompt and answer**. It is a "second opinion," not a private one. When a prompt must stay out of the harness's request, use [`private`](/plugin/commands/private) instead.

## Ask

<Tabs>
  <Tab title="Claude Code">
    ```text theme={"theme":{"light":"github-light","dark":"github-dark"}}
    /nito:ask [--model <id>] [--file <path>]... [--history|--no-history] [--web-search off|auto|on] [--web-fetch=off|on] [--] <prompt>
    ```
  </Tab>

  <Tab title="Codex">
    ```text theme={"theme":{"light":"github-light","dark":"github-dark"}}
    $nito:ask [--model <id>] [--file <path>]... [--history|--no-history] [--web-search off|auto|on] [--web-fetch=off|on] [--] <prompt>
    ```
  </Tab>
</Tabs>

The result is labeled with the environment, model, privacy level, and context scope, for example `Nito · Hosted · z-ai/glm-5.2 · private · Prompt only`.

### Options

| Option | What it does |
| :- | :- |
| `--model <id>` | Use a specific model for this call. Without it, `ask` uses your saved default. See [Choosing a Model](/plugin/commands/choosing-a-model). |
| `--file <path>` | Attach a local file. Repeatable. |
| `--history` / `--no-history` | Include or exclude conversation context for this one call. Authoritative over the [Context](/plugin/commands/context-and-mode) preference. |
| `--web-search off\|auto\|on` | Control web search. See [Web Search](/plugin/features/server-side-tools/web-search). |
| `--web-fetch=off\|on` | Control URL retrieval. See [Web Fetch](/plugin/features/server-side-tools/web-fetch). |
| `--` | End of options; everything after is the prompt. |

### What the Harness Can See

By default `ask` sends **only your prompt** to Nito, and the exchange stays in your harness chat where the harness model can see it. Turning [Context](/plugin/commands/context-and-mode) on lets `ask` include a short harness-generated summary of the task; nothing is sent that you did not enable.

## Continue

`continue` reuses the **immediately preceding** Nito exchange (that prompt and answer) plus your new prompt, and nothing else from the conversation. It is for following up without re-pasting.

<Tabs>
  <Tab title="Claude Code">
    ```text theme={"theme":{"light":"github-light","dark":"github-dark"}}
    /nito:continue [--model <id>] [--file <path>]... <prompt>
    ```
  </Tab>

  <Tab title="Codex">
    ```text theme={"theme":{"light":"github-light","dark":"github-dark"}}
    $nito:continue [--model <id>] [--file <path>]... <prompt>
    ```
  </Tab>
</Tabs>

If the preceding Nito exchange is unavailable or ambiguous, `continue` does nothing and asks you to use `ask` instead. Passing `--model` overrides the model for the follow-up.

## Next

Pick which model answers with [Choosing a Model](/plugin/commands/choosing-a-model), or keep a question out of your harness entirely with [`private`](/plugin/commands/private).


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.