***

id: 43b15040-3108-4b7f-83d3-ac23153e4106
hide\_title: false
slug: /reference/amazon-bedrock/prompt
title: prompt
description: >-
Establish the set of rules and instructions for the Amazon Bedrock agent
through a prompt.
max-toc-depth: 3
----------------

[prompt-properties]: #properties

## **Properties**

<ParamField path="amazon_bedrock.prompt" type="object" required={true} toc={true}>
  An object that accepts the [`prompt properties`][prompt-properties].
</ParamField>

The `prompt` property accepts one of the following objects:

<Tabs>
  <Tab title="Regular Prompt">
    <ParamField path="prompt.text" type="string" required={true} toc={true}>
      The instructions to send to the agent.
    </ParamField>

    <ParamField path="prompt.temperature" type="number" default="1.0" toc={true}>
      Randomness setting. Float value between 0.0 and 1.5. Closer to 0 will make the output less random.
    </ParamField>

    <ParamField path="prompt.top_p" type="number" default="1.0" toc={true}>
      Randomness setting. Alternative to temperature. Float value between 0.0 and 1.0.
    </ParamField>

    <ParamField path="prompt.confidence" type="number" default="0.6" toc={true}>
      Threshold to fire a speech-detect event at the end of the utterance.
    </ParamField>

    <ParamField path="voice_id" type="string" default="matthew" toc={true}>
      The voice the Amazon Bedrock agent will use during the interaction.
      Possible Values: `tiffany`, `matthew`, `amy`, `lupe`, `carlos`
    </ParamField>
  </Tab>

  <Tab title="POM Prompts">
    <ParamField path="prompt.pom" type="object[]" required={true} toc={true}>
      An array of objects that defines the prompt object model (POM) for the agent. The POM is a structured data format for organizing and rendering prompts that are clearly structured and easy to understand. By breaking prompts into sections, users can manage each section independently and then combine them into a single cohesive prompt. SignalWire renders the POM into a markdown document before sending it to the LLM. If the `text` parameter is present while using `pom`, the `pom` prompt will be used instead of `text`.
    </ParamField>

    <Indent>
      <ParamField path="pom[].title" type="string" toc={true}>
        The title of the section. Will be a heading in the rendered prompt.
      </ParamField>

      <ParamField path="pom[].body" type="string" toc={true}>
        The body of the section. This will be a paragraph in the rendered prompt.
        Required if `bullets` is not present.
      </ParamField>

      <ParamField path="pom[].bullets" type="string[]" toc={true}>
        An array of strings that represent the bullets of the section.
        This will be a list of short statements/rules in the rendered prompt.
        Optional if `body` is present.
      </ParamField>

      <ParamField path="pom[].subsections" type="object[]" toc={true}>
        An array of section objects allowing users to nest sections within sections. Each subsection accepts the same properties as a top-level POM section (`title`, `body`, `bullets`, `numbered`, `numberedBullets`).
      </ParamField>

      <ParamField path="pom[].numbered" type="boolean" default="false" toc={true}>
        If `true`, the section will be numbered in the rendered prompt.
      </ParamField>

      <ParamField path="pom[].numberedBullets" type="boolean" default="false" toc={true}>
        If `true`, the bullets will be numbered in the rendered prompt.
      </ParamField>
    </Indent>

    <ParamField path="prompt.temperature" type="number" default="1.0" toc={true}>
      Randomness setting. Float value between 0.0 and 1.5. Closer to 0 will make the output less random.
    </ParamField>

    <ParamField path="prompt.top_p" type="number" default="1.0" toc={true}>
      Randomness setting. Alternative to temperature. Float value between 0.0 and 1.0.
    </ParamField>

    <ParamField path="prompt.confidence" type="number" default="0.6" toc={true}>
      Threshold to fire a speech-detect event at the end of the utterance.
    </ParamField>

    <ParamField path="voice_id" type="string" default="matthew" toc={true}>
      The voice the Amazon Bedrock agent will use during the interaction.
      Possible Values: `tiffany`, `matthew`, `amy`, `lupe`, `carlos`
    </ParamField>

    #### POM example

    <CodeBlocks>
      <CodeBlock title="YAML">
        ```yaml
        version: 1.0.0
        sections:
          main:
            - amazon_bedrock:
                prompt:
                  text: "Prompt is defined in pom"
                  pom:
                    - title: "Agent Personality"
                      body: "You are a friendly and engaging assistant. Keep the conversation light and fun."
                      subsections:
                        - title: "Personal Information"
                          numberedBullets: true
                          bullets:
                            - "You are a AI Agent"
                            - "Your name is Frank"
                            - "You work at SignalWire"
                    - title: "Task"
                      body: "You are to ask the user a series of questions to gather information."
                      bullets:
                        - "Ask the user to provide their name"
                        - "Ask the user to provide their favorite color"
                        - "Ask the user to provide their favorite food"
                        - "Ask the user to provide their favorite movie"
                        - "Ask the user to provide their favorite TV show"
                        - "Ask the user to provide their favorite music"
                        - "Ask the user to provide their favorite sport"
                        - "Ask the user to provide their favorite animal"
        ```
      </CodeBlock>

      <CodeBlock title="JSON">
        ```json
        {
          "version": "1.0.0",
          "sections": {
            "main": [
              {
                "amazon_bedrock": {
                  "prompt": {
                    "text": "Prompt is defined in pom",
                    "pom": [
                      {
                        "title": "Agent Personality",
                        "body": "You are a friendly and engaging assistant. Keep the conversation light and fun.",
                        "subsections": [
                          {
                            "title": "Personal Information",
                            "numberedBullets": true,
                            "bullets": [
                              "You are a AI Agent",
                              "Your name is Frank",
                              "You work at SignalWire"
                            ]
                          }
                        ]
                      },
                      {
                        "title": "Task",
                        "body": "You are to ask the user a series of questions to gather information.",
                        "bullets": [
                          "Ask the user to provide their name",
                          "Ask the user to provide their favorite color",
                          "Ask the user to provide their favorite food",
                          "Ask the user to provide their favorite movie",
                          "Ask the user to provide their favorite TV show",
                          "Ask the user to provide their favorite music",
                          "Ask the user to provide their favorite sport",
                          "Ask the user to provide their favorite animal"
                        ]
                      }
                    ]
                  }
                }
              }
            ]
          }
        }
        ```
      </CodeBlock>
    </CodeBlocks>

    #### Rendered prompt

    The above POM example will render the following markdown prompt:

    ```markdown
    ## Agent Personality

    You are a friendly and engaging assistant. Keep the conversation light and fun.

    ### Personal Information

    1. You are a AI Agent
    2. Your name is Frank
    3. You work at SignalWire

    ## Task

    You are to ask the user a series of questions to gather information.

    - Ask the user to provide their name
    - Ask the user to provide their favorite color
    - Ask the user to provide their favorite food
    - Ask the user to provide their favorite movie
    - Ask the user to provide their favorite TV show
    - Ask the user to provide their favorite music
    - Ask the user to provide their favorite sport
    - Ask the user to provide their favorite animal
    ```
  </Tab>
</Tabs>

## **Variable Expansion**

Use the following syntax to expand variables into your prompt.

<ParamField path="${call_direction}" type="string" toc={true}>
  Inbound or outbound.
</ParamField>

<ParamField path="${caller_id_number}" type="string" toc={true}>
  The caller ID number.
</ParamField>

<ParamField path="${local_date}" type="string" toc={true}>
  The local date.
</ParamField>

<ParamField path="${spoken_date}" type="string" toc={true}>
  The spoken date.
</ParamField>

<ParamField path="${local_time}" type="string" toc={true}>
  The local time.
</ParamField>

<ParamField path="${time_of_day}" type="string" toc={true}>
  The time of day.
</ParamField>

<ParamField path="${supported_languages}" type="string" toc={true}>
  A list of supported languages.
</ParamField>

<ParamField path="${default_language}" type="string" toc={true}>
  The default language.
</ParamField>