Prompty Specification

The Prompty yaml file spec can be found here. Below you can find a brief description of each section and the attributes within it.

Prompty description attributes:

Property	Type	Description	Required
`name`	string	Name of the prompty	Yes
`description`	string	Description of the prompty	Yes
`version`	string	Version number	No
`authors`	array	List of prompty authors	No
`tags`	tags	Categorization tags	No

Input/Output Specifications

Inputs

The inputs object defines the expected input format for the prompty:

inputs:
  type: object
  description: "Input specification"

Outputs

The outputs object defines the expected output format:

outputs:
  type: object
  description: "Output specification"

Template Engine

Currently supports:

jinja2 (default) - Jinja2 template engine for text processing

Model Configuration

The model section defines how the AI model should be configured and executed.

model:
  api: chat
  configuration:
    # model-specific configuration
  parameters:
    # execution parameters
  response: first

Model API Types

chat (default) - For chat-based interactions
completion - For text completion tasks

Response Types

This determines whether the full (raw) response or just the first response in the choice array is returned.

first (default) - Returns only the first response
all - Returns all response choices

Model Providers

Azure OpenAI Configuration

configuration:
  type: azure_openai
  api_key: ${env:OPENAI_API_KEY}
  api_version: "2023-05-15"
  azure_deployment: "my-deployment"
  azure_endpoint: "https://my-endpoint.openai.azure.com"

OpenAI Configuration

configuration:
  type: openai
  name: "gpt-4"
  organization: "my-org"

MaaS Configuration

configuration:
  type: azure_serverless
  azure_endpoint: "https://my-endpoint.azureml.ms"

Model Parameters

Common parameters that can be configured for model execution:

Parameter	Type	Description
`response_format`	object	An object specifying the format that the model must output.
`seed`	integer	For deterministic sampling. This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same `seed` and parameters should return the same result. Determinism is not guaranteed, and you should refer to the `system_fingerprint` response parameter to monitor changes in the backend.
`max_tokens`	integer	The maximum number of tokens that can be generated in the chat completion.
`temperature`	number	Sampling temperature (0-1)
`frequency_penalty`	number	Penalty for frequent tokens
`presence_penalty`	number	Penalty for new tokens
`top_p`	number	Nucleus sampling probability
`stop`	array	Sequences to stop generation

Setting to { "type": "json_object" } enables JSON mode, which guarantees the message the model generates is valid JSON.

Sample Prompty

---
name: ExamplePrompt
description: A prompt that uses context to ground an incoming question
authors:
  - Seth Juarez
model:
  api: chat
  configuration:
    type: azure_openai
    azure_endpoint: ${env:AZURE_OPENAI_ENDPOINT}
    azure_deployment: <your-deployment>
    api_version: 2024-07-01-preview
  parameters:
    max_tokens: 3000
sample:
  firstName: Seth
  context: >
    The Alpine Explorer Tent boasts a detachable divider for privacy,
    numerous mesh windows and adjustable vents for ventilation, and
    a waterproof design. It even has a built-in gear loft for storing
    your outdoor essentials. In short, it's a blend of privacy, comfort,
    and convenience, making it your second home in the heart of nature!
  question: What can you tell me about your tents?
---

system:
You are an AI assistant who helps people find information. As the assistant,
you answer questions briefly, succinctly, and in a personable manner using
markdown and even add some personal flair with appropriate emojis.

# Customer
You are helping {{firstName}} to find answers to their questions.
Use their name to address them in your responses.

# Context
Use the following context to provide a more personalized response to {{firstName}}:
{{context}}

user:
{{question}}

Want to Contribute To the Project?