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

# NLU Command Adapter

> How the NLUCommandAdapter Works

## How the NLUCommandAdapter Works

The `NLUCommandAdapter` uses the classic way to start flows, such as using predicted intents by an intent classifier.
It looks at the predicted intent from the [intent classifier](/docs/reference/config/components/nlu-components#intent-classifiers)
and tries to find a flow with a corresponding [NLU trigger](/docs/reference/primitives/starting-flows#nlu-trigger) defined.
If a flow has a NLU trigger matching the predicted intent and the confidence is larger than the given threshold defined
in the NLU trigger, the `NLUCommandAdapter` will return a `StartFlow` command to begin the corresponding flow.

## Using the NLUCommandAdapter

To use this component in your assistant, add the `NLUCommandAdapter` to your NLU pipeline in the `config.yml` file.
You also need to have an [intent classifier](/docs/reference/config/components/nlu-components#intent-classifiers)
listed in your NLU pipeline. Read more about the `config.yml` file [here](/docs/reference/config/overview).

```yaml title="config.yml" theme={null}
pipeline:
# - ...
  - name: NLUCommandAdapter
# - ...
```

## When to use the NLUCommandAdapter

We recommend to use the `NLUCommandAdapter` in two scenarios:

* You want to use NLU data containing intent and examples along with the CALM paradigm. Using the `NLUCommandAdapter` you can initiate
  a flow based on a predicted intent, given you already have a solid intent classifier in place. Once the flow is initiated, the business
  logic would be executed as usual in the CALM paradigm with commands predicted by the `LLMCommandGenerator` and policies predicting the next best action.
* You want to minimize the costs by not making an API call to the LLM each time. The `NLUCommandAdapter` does not make
  any API call to a LLM compared to the `LLMCommandGenerator`. Using the `NLUCommandAdapter` saves some costs.
  Make sure you have a solid intent classifier in place when using the `NLUCommandAdapter`;
  otherwise, incorrect flows will begin.

<Info>
  **New in 3.12**

  If you are using a LLM-based command generator alongside the `NLUCommandAdapter` in the config pipeline, note that
  both the LLM-based command generator and the `NLUCommandAdapter` can now issue commands at any given
  conversation turn. To enable this behaviour, you should set the
  `minimize_num_calls` boolean parameter to `false` in the
  [LLM-based command generator configuration](/docs/reference/config/components/llm-command-generators#interaction-with-other-types-of-command-generators).
</Info>

## Customization

To restrict the length of user messages, you can set the `user_input.max_characters` (default value is 420 characters).

```yaml title="config.yml" {3} theme={null}
pipeline:
  - name: NLUCommandAdapter
    user_input:
      max_characters: 420
```


## Related topics

- [Tracing](/docs/reference/integrations/tracing.md)
- [Starting Flows](/docs/reference/primitives/starting-flows.md)
- [Slots](/docs/reference/primitives/slots.md)
- [LLM Command Generators](/docs/reference/config/components/llm-command-generators.md)
- [Rasa Pro Change Log](/docs/reference/changelogs/rasa-pro-changelog.md)