> ## 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.

# Policy Overview

> In Rasa, policies are the components responsible for dialogue management.

In Rasa, policies are the components responsible for dialogue management.
In a CALM-based assistant, the [FlowPolicy](/docs/reference/config/policies/flow-policy) is responsible for executing your business logic.
If you're building an NLU-based assistant, you can read about the relevant
policies [here](https://legacy-docs-oss.rasa.com/docs/rasa/policies).

<Note>
  **Policies in CALM**

  [Dialogue Understanding](/docs/learn/concepts/dialogue-understanding) in CALM accounts
  for the context of a conversation, so the role of the dialogue manager is simpler than it is in
  an NLU-based assistant.
</Note>

You can customize the policies your assistant uses by specifying the `policies`
key in your project's `config.yml`.
In most cases, the default configuration should meet your needs, customization is
there for advanced use-cases.

There are different policies to choose from, and you can include
multiple policies in a single configuration. Here's an example of
what a list of policies might look like:

<Tabs>
  <Tab title="Rasa Pro <=3.7.x">
    ```yaml title="config.yml" theme={null}
    # ...

    policies:
      - name: FlowPolicy
      - name: rasa_plus.ml.EnterpriseSearchPolicy
    ```
  </Tab>

  <Tab title="Rasa Pro >=3.8.x">
    ```yaml title="config.yml" theme={null}
    # ...

    policies:
      - name: FlowPolicy
      - name: EnterpriseSearchPolicy
    ```
  </Tab>
</Tabs>

## Action Selection

At every turn, each policy defined in your configuration gets a chance to
predict a next [action](/docs/reference/primitives/actions) with a certain confidence level.
A policy can also decide not to predict any action.
The policy that predicts with the highest confidence decides the assistant's next action.

<Note>
  **Maximum number of predictions**

  By default, when using CALM assistant (or routing to it, in case of
  [coexistence of NLU-based and CALM systems](/docs/pro/calm-with-nlu/migrating-from-nlu))
  the maximum number of next action predictions after each user message is 1000.
  To update this value, you can set the environment variable `MAX_NUMBER_OF_PREDICTIONS_CALM`
  to the desired number of maximum predictions. For an NLU-based assistant the maximum is
  set in the environment variable `MAX_NUMBER_OF_PREDICTIONS` with a default of 10.
</Note>


## Related topics

- [Overview](/docs/reference/config/overview.md)
- [PII Management Overview](/docs/reference/config/pii-management/overview.md)
- [Installation Overview](/docs/studio/installation/overview.md)
- [How Does Coexistence Work?](/docs/pro/calm-with-nlu/coexistence.md)