SDK labs: JavaScript and Python

JavaScript

npm install @typesafe-ai/sdk

The SDK is like a coffee machine. You don't grind the beans and boil the water yourself. You press a button. Same idea here: you call ready-made functions instead of writing raw HTTP requests.

import { choice, noul, score, TypeSafeClient } from "@typesafe-ai/sdk";

const client = new TypeSafeClient();
const response = await client.systemOne({
  state: { document: "I was charged twice. Please fix this ASAP." },
  questions: {
    category: choice("What is this ticket about?", {
      billing: "Payments, charges, refunds",
      technical: "Bugs or integrations",
      other: "Anything else",
    }),
    urgent: noul("Does this convey urgency?"),
    frustration: score("How frustrated is the customer?", [
      "Calm",
      "Frustrated",
      "Very angry",
    ]),
  },
});

That call sends one support ticket and three questions. A Choice picks the category. A Noul checks urgency. A Score rates frustration. One call to System One, three answers back.

Do not set dangerouslyAllowBrowser: true on public academy pages (JS SDK).

Python

pip install typesafe-sdk

from typesafe_sdk import Choice, Noul, Score, TypeSafeClient

with TypeSafeClient() as client:
    response = client.system_one(
        state={"document": "I was charged twice. Please fix this ASAP."},
        questions={
            "billing": Noul(instructions="Is this ticket about billing?"),
            "tone": Choice(
                instructions="What is the customer's tone?",
                criteria={"calm": None, "frustrated": None, "angry": None},
            ),
            "urgency": Score(
                instructions="How urgent is this ticket?",
                criteria=["can wait", "this week", "today"],
            ),
        },
    )

Same idea in Python. One ticket, three questions, one system_one call.

Models: prefer jev-latest. It's an alias, so the exact model behind it can change. Pin jev-1.13.0 when you need stable thresholds (Models). Optional extra: the Agent skill.

Next

Capstone labs and the use-case gallery.