Crate jev_sdk 

jev-sdk

A Rust client for the TypeSafe AI System One API.

Jev is a System One model: instead of generating text, it evaluates typed questions against a state and returns structured decisions with calibrated probabilities and confidence — fast enough to sit inside ordinary code paths.

Quick start

use jev_sdk::{Choice, Noul, Question, Score, TypeSafeClient};

// Reads TYPESAFE_API_KEY from the environment.
let client = TypeSafeClient::from_env()?;

let ticket = "Hi, I've been trying to connect my Stripe account for 3 days \
              and it keeps failing. I'm losing sales. Please help ASAP.";

let response = client
    .system_one(
        ticket,
        [
            (
                "department",
                Question::from(Choice::new(
                    "Which team should handle this",
                    [
                        ("billing", Some("Payment or subscription issues")),
                        ("technical", Some("Bugs or integration problems")),
                        ("sales", Some("Pricing or account questions")),
                    ],
                )),
            ),
            (
                "frustration",
                Question::from(Score::new(
                    "How frustrated the customer appears",
                    ["Calm, just stating facts", "Frustrated but civil", "Very angry"],
                )),
            ),
            (
                "is_urgent",
                Question::from(Noul::new("The message conveys urgency or time-sensitivity")),
            ),
        ],
    )
    .await?;

let department = response.choice("department").unwrap();
println!("route to {} (confidence {:.2})", department.choice, department.confidence);
println!("frustration: {:.2}", response.score("frustration").unwrap().score);
println!("urgent: {:.3}", response.noul("is_urgent").unwrap().noul);

Configuration

Setting	Environment variable	Default
API key	TYPESAFE_API_KEY	— (required)
Model	TYPESAFE_DEFAULT_MODEL	jev-latest
Base URL	TYPESAFE_BASE_URL	https://api.typesafe.ai

Use TypeSafeClient::builder for timeouts, retry policy, extra headers, or a custom reqwest client. See RetryPolicy for retry behavior (transient 429/529 responses are retried with exponential backoff by default).

Structs

ApiError

An unsuccessful HTTP response from the TypeSafe API.

Choice

Pick one option from a defined set.

ChoiceAnswer

Answer to a choice question.

ClientBuilder

Builds TypeSafeClients.

ListModelsResponse

The response to list_models.

ModelCard

A model available to the account.

Noul

A yes/no question. The answer is the probability that the answer is “yes”.

NoulAnswer

Answer to a noul question: the probability that the answer is “yes”.

NoulCriteria

Optional descriptions of the two outcomes of a Noul question.

RetryPolicy

Controls how failed requests are retried.

Score

Rate the state along an ordered rubric.

ScoreAnswer

Answer to a score question.

SystemOneResponse

The response to a system_one evaluation.

TypeSafeClient

An asynchronous client for the TypeSafe AI System One API.

Usage

Token usage for a request.

Enums

Answer

One typed answer. The variant matches the type of the question asked.

Error

Errors returned by this SDK.

Question

A typed question to evaluate against the state.

Constants

DEFAULT_BASE_URL

Default API root.

DEFAULT_MODEL

Default model used when none is configured.

DEFAULT_TIMEOUT

Default request timeout.

ENV_API_KEY

Environment variable for the API key.

ENV_BASE_URL

Environment variable for the API root.

ENV_DEFAULT_MODEL

Environment variable for the default model.

Traits

IntoContent

Any content that can be sent as part of a question: a plain string or arbitrary JSON structure.

IntoDescription

An optional description for a Choice option.