AI quickstart

Make your first model call crash-safe: add a durable AI step, point it at your provider, and watch its spend, tokens, and latency land in the console.

Turn a model call into a durable step, trigger it, and watch its spend, tokens, and latency roll up in the console's AI view.

This picks up where the Quickstart leaves off - it assumes you already have a project, an API key, and a connected runner. If you don't, start there first.

1. Add an AI step

ctx.step.ai.generate makes a model call a durable step: its result is recorded once under the step id, so a retry after a crash returns the saved result instead of calling - and paying for - the model again.

runner.ts

import { defineWorkflow } from "@duraton/sdk";

export const triageTicket = defineWorkflow<{ subject: string }>({
  name: "ticket.created",
  handler: async (ctx) => {
    const { text } = await ctx.step.ai.generate("classify", {
      model: "claude-opus-4-8",
      prompt: `Classify this ticket: ${ctx.event.data.subject}`,
    });
    return { text };
  },
});

runner.py

from duraton import GenerateOptions, define_workflow
from duraton.context import StepContext


async def triage(ctx: StepContext) -> object:
    result = await ctx.step.ai.generate("classify", GenerateOptions(
        model="claude-opus-4-8",
        prompt=f"Classify this ticket: {ctx.event.data['subject']}",
    ))
    return {"text": result.text}


triage_ticket = define_workflow("ticket.created", triage)

runner.go

type Ticket struct {
	Subject string `json:"subject"`
}

var triageTicket = duraton.DefineWorkflow(duraton.WorkflowDefinition{
	Name: "ticket.created",
	Handler: func(c *duraton.Context) (any, error) {
		ticket, err := duraton.EventData[Ticket](c)
		if err != nil {
			return nil, err
		}
		res, err := duraton.Generate(c, "classify", duraton.GenerateOptions{
			Model:  "claude-opus-4-8",
			Prompt: "Classify this ticket: " + ticket.Subject,
		})
		if err != nil {
			return nil, err
		}
		return map[string]string{"text": res.Text}, nil
	},
})

2. Set your provider key

The built-in provider is Anthropic. Your runner makes the model call, so the key stays with your runner - Duraton meters tokens but never sees your key, your prompt, or the response. Set it in the runner's environment:

export ANTHROPIC_API_KEY="sk-ant-..."

Duraton is bring-your-own-keys. Omit apiKey on the call and the provider SDK reads its conventional env var (ANTHROPIC_API_KEY); pass apiKey per call to override it. Either way the key is used for that one call and is never recorded in the run history.

3. Trigger it and watch spend land

Send the event your workflow listens for - with the SDK client, or over the REST API:

import { createClient } from "@duraton/sdk";

const duraton = createClient({
  url: process.env.DURATON_URL!,
  apiKey: process.env.DURATON_API_KEY,
});

await duraton.events.send({
  name: "ticket.created",
  data: { subject: "Refund not received" },
});

import os

from duraton.client import AsyncDuratonClient, SendEventInput

async with AsyncDuratonClient(
    url=os.environ["DURATON_URL"],
    api_key=os.environ.get("DURATON_API_KEY"),
) as dx:
    await dx.events.send(SendEventInput(
        name="ticket.created",
        data={"subject": "Refund not received"},
    ))

import (
	"context"
	"encoding/json"

	"duraton.dev/sdk-go/client"
)

dx := client.New(client.Options{})

_, err := dx.Events.Send(context.Background(), client.SendEventInput{
	Name: "ticket.created",
	Data: json.RawMessage(`{"subject":"Refund not received"}`),
})

curl -X POST "$DURATON_URL/events" \
  -H "Authorization: Bearer $DURATON_API_KEY" \
  -d '{"name":"ticket.created","data":{"subject":"Refund not received"}}'

As the AI step runs, open the AI view in the console. Once the first call is recorded it fills in: total spend and tokens for the window, average latency, cache hit rate, and breakdowns of cost by hour, by model, and by workflow.

1. Add an AI step

2. Set your provider key

3. Trigger it and watch spend land

Read it over MCP

Next steps

AI steps

AI agents

Cost controls

Evals

On this page