Create a prompt

Define a prompt object in Arize’s Prompt Playground.

Write your prompt in the Playground, configure your model, parameters, and functions (see Workflow on this page). After, test your prompt on real data, then save to Prompt Hub once you’re satisfied.

Workflow

By Arize Skills

Coming soon!

By Alyx

1. Create a new playground

   Navigate to Playgrounds and click New Playground to create a new playground. You can also start from the home page, and Alyx will create a playground for you.

2. Choose a provider or model

   If you have not already added one, select your LLM provider and model in the Playground before you draft or run. Options include OpenAI, Azure, Vertex AI, AWS Bedrock, and others.

   
3. Draft and refine your prompt with Alyx

   Describe your use case in plain language in the Alyx panel on the right. Include details about what your prompt should capture, desired output format, tone, length, etc. Alyx will wire them in automatically.

   For example, you could say:

   Create a prompt for a customer support agent that handles returns and escalations. Use {customer_input} and {order_id} as variables.

   Give feedback on Alyx’s drafts and keep iterating until you are satisfied with the prompt.
4. Try the prompt in the Playground

   If you added a dataset or test cases, run the prompt over rows to check outputs. If you skipped that step, exercise the prompt with manual inputs or keep iterating with Alyx. For a fuller testing workflow, see Test a prompt.

For more patterns, see the Playground Agent.

By UI

1. Create a new playground

   Navigate to Playgrounds and click New Playground to create a new playground.

2. Choose provider and model

   Select your LLM provider and model from OpenAI, Azure, Vertex AI, AWS Bedrock, and others.

   
3. Add prompt messages

   Add any system, user, or assistant messages. Use curly braces around names (for example, {destination}) for template variables filled dynamically at run time.
4. Add function calling (optional)

   Click Functions and paste tool definitions as JSON. Function descriptions must be in JSON format. To add multiple functions, add entries to the JSON list with the proper attributes. Set Function Selection to give the model instructions on when to invoke each function.

   
5. Set invocation parameters

   Click Params to configure temperature, max completion tokens, frequency penalty, and more. Different LLM providers expose different hyperparameters that affect the output.

By Code

Create a prompt and initial version from code. Use the reference for your stack:

Python SDK

Full options (providers, invocation params, tools): Create a Prompt in the Python v8 Prompts API.

from arize._generated.api_client.models.input_variable_format import InputVariableFormat
from arize._generated.api_client.models.llm_provider import LlmProvider
from arize._generated.api_client.models.llm_message import LLMMessage

prompt = client.prompts.create(
    space="your-space-name-or-id",
    name="customer-support-agent",
    description="Handles returns and escalations",
    commit_message="Initial version",
    input_variable_format=InputVariableFormat.F_STRING,
    provider=LlmProvider.OPENAI,
    model="gpt-4o",
    messages=[
        LLMMessage(role="system", content="You are a support agent for {company}."),
        LLMMessage(role="user", content="{ticket_text}"),
    ],
)

print(prompt.id, prompt.name)

TypeScript SDK

Full options: Create a Prompt in the TypeScript Prompts API.

import { createPrompt } from "@arizeai/ax-client";

const prompt = await createPrompt({
  space: "my-space",
  name: "customer-support",
  description: "Handles returns and escalations",
  version: {
    commitMessage: "Initial version",
    inputVariableFormat: "f_string",
    provider: "openAI",
    model: "gpt-4o",
    messages: [
      { role: "system", content: "You are a support agent for {company_name}." },
      { role: "user", content: "{ticket_text}" },
    ],
  },
});

CLI

Flags and message JSON shape: ax prompts create in the CLI reference.

ax prompts create \
  --name "customer-support-agent" \
  --space sp_abc123 \
  --provider openAI \
  --input-variable-format f_string \
  --messages '[{"role":"system","content":"You are a support agent for {company}."},{"role":"user","content":"{ticket_text}"}]' \
  --model gpt-4o \
  --description "Handles returns and escalations" \
  --commit-message "Initial version"

Next up

With a draft in the Playground, you are ready to Test a prompt on datasets or traces and score runs with evaluators.