OpenJev
Jev-inspired decision API with a local mock server and Python and TypeScript SDKs for prototyping; real inference is planned.
Overview
OpenJev (xingwudao) is an independent, Jev-inspired System One decision API with choice, score and noul primitives. The first runnable release includes a local mock HTTP API, Python and JavaScript/TypeScript SDKs, schema validation and end-to-end tests, for prototyping ticket routing, classification, scoring and guardrail workflows.
The README is explicit that mock probabilities are synthetic and do not measure truth, risk or model quality, that only the openjev-mock model is supported, and that real inference is planned. The server binds to loopback without authentication and is not production infrastructure. It includes a Chinese summary and an explanation of what Jev is used for. It suits developers who want to build integration code against the decision interface before choosing a model.
What the decision model decides
| Type | Question | Answers |
|---|---|---|
| Choice | Which team should handle this ticket? (mock)(inferred) | — |
| Score | Risk or severity score (mock)(inferred) | — |
| Yes / no | Does this need a guardrail? (mock)(inferred) | yes · no |
How it works
- Start the mock server locally.
- Call /v1/system_one from Python or TypeScript SDKs.
- Build and test branching logic against synthetic probabilities.
- Swap in a real model when available.
Strengths
- Lets you build and test integrations without a key.
- SDKs in two languages with tests.
Caveats
- Mock only; probabilities are synthetic.
- No real inference yet; not production-ready.
Getting started
python3 -m venv .venv && .venv/bin/python -m pip install -r requirements.txt -e sdk/python && .venv/bin/python -m server --port 8000At a glance
- Type
- Library
- Status
- prototype
- Language
- Python
- Runtime
- local mock server, SDKs
- Has code
- yes
- Has tests
- yes
- Stars
- 5
Use cases and integrations
How curated lists describe it
OpenJev: an independent Jev-inspired System One decision API based on TypeSafe.ai concepts. Choice, score and noul primitives, local mock server, Python and TypeScript SDKs. Real inference planned; not affiliated with TypeSafe AI.
wh000wh000/awesome-jev-liveOpenJev: an independent Jev-inspired System One decision API based on TypeSafe.ai concepts. Choice, score and noul primitives, local mock server, Python and TypeScript SDKs. Real inference planned; not affiliated with TypeSafe AI.
hellogumbo/awesome-jev
Frequently asked questions
Does it run a real model?
No, only a mock with synthetic probabilities; real inference is planned.
What is it good for now?
Prototyping and testing integration code against the decision interface.
Is it affiliated with TypeSafe?
No.
Related projects
Python tap SDK for Android, iOS and HarmonyOS where multimodal Jev picks a numbered on-screen candidate and the SDK taps it.
LibrarybetaPythonMSW handlers that mock the Jev API for the official TypeSafe SDK, making Jev-driven tests deterministic without keys or credits.
ServicebetaTypeScriptHack Club's lightweight LLM proxy, listed for adding a Jev forwarding endpoint behind its auth, limits and usage logging.
MCP serverproductionTypeScriptAx, a DSPy-style multi-language LLM framework, listed with a native TypeSafe provider for Boolean, Choice, Score and raw Jev questions.
PluginproductionTypeScriptEffect TypeScript library whose @effect/ai-typesafe package maps classify, probability and rating operations to Jev questions.
ProjectproductionTypeScriptPHP monorepo whose Polyglot module adds a TypeSafe driver for typed Noul, Choice and Score decisions alongside LLM inference.
ProjectproductionPHP