Skip to content
Metatext Decide — typed decisions for your agents, with a probability on every answerGet early access
Metatext
Sign Up

jev\_dart

Pure Dart client for Jev that sends state and typed questions from CLI, server or Flutter apps, tuned for low overhead.

CLIbetaMIT
View on GitHub ↗

Overview

jev_dart is a Dart package for calling TypeSafe's Jev. You send state and typed noul, choice or score questions and get structured answers your code can branch on. It has no Flutter dependency, so it works from command-line tools, servers and Flutter apps alike.

The README emphasises client-side speed and memory use: one long-lived HTTP client reuses its HTTP/2 connection, request JSON is encoded straight to UTF-8 bytes and reused across retries, and responses are parsed directly from bytes with Crimson on VM, mobile and desktop, with a standard dart:convert fallback. It reads TYPESAFE_API_KEY or JEV_API_KEY, including from .env in tests and examples.

What the decision model decides

TypeQuestionAnswers
ChoiceWhat is this ticket about?billing · technical · other

How it works

  1. Add jev_dart to pubspec dependencies and set TYPESAFE_API_KEY
  2. Create a TypeSafeClient
  3. Call systemOne with state and typed questions
  4. Read typed answers and close the client

Strengths

  • Works in CLI, server and Flutter
  • Lower client-side encoding overhead and connection reuse
  • Typed helpers for choice, noul and score

Caveats

  • Community SDK; API coverage beyond the example not detailed
  • Requires a TypeSafe key

Getting started

Add `jev_dart` to pubspec.yaml, set TYPESAFE_API_KEY, and call `TypeSafeClient().systemOne(state: ..., questions: ...)`.

At a glance

Type
CLI
Status
beta
Language
Dart
Runtime
Dart VM, Flutter, CLI
Has code
yes
Has tests
yes
License
MIT

Use cases and integrations

How curated lists describe it

Frequently asked questions

Does it need Flutter?

No, it is pure Dart and works in CLIs and servers too.

What makes it different?

It focuses on low client overhead: connection reuse, direct byte encoding and byte-level JSON parsing.

Which key does it read?

TYPESAFE_API_KEY or JEV_API_KEY.

Related projects

  • ai

    Hack Club's lightweight LLM proxy, listed for adding a Jev forwarding endpoint behind its auth, limits and usage logging.

    MCP serverproductionTypeScript
  • Ax

    Ax, a DSPy-style multi-language LLM framework, listed with a native TypeSafe provider for Boolean, Choice, Score and raw Jev questions.

    PluginproductionTypeScript
  • Effect

    Effect TypeScript library whose @effect/ai-typesafe package maps classify, probability and rating operations to Jev questions.

    ProjectproductionTypeScript
  • instructor-php

    PHP monorepo whose Polyglot module adds a TypeSafe driver for typed Noul, Choice and Score decisions alongside LLM inference.

    ProjectproductionPHP
  • jev-mcp (freepik-company)

    Go MCP server giving any client typed Jev decisions via OpenRouter or TypeSafe, with validated answers and no paid retries.

    MCP serverproductionGo
  • llm-typesafe

    Simon Willison's LLM plugin that adds Jev as a model for yes/no, choice and score questions from the command line.

    PluginproductionPython