0312GitHub

System One Harness

Turns a decision model into an agent loop: it compiles the environment's finite action space into typed questions, gates each step by confidence, and records the trace.

Open source ↗ github.comcost-time-

HarnessRouter/SystemOneHarnessREADME ↗

The harness for System One models.</sub>
</h1>

<p align="center">
  <a href="https://github.com/HarnessRouter/SystemOneHarness" title="Star System One Harness on GitHub"><img src="https://img.shields.io/github/stars/HarnessRouter/SystemOneHarness?style=flat&amp;logo=github&amp;logoColor=white&amp;label=Stars&amp;labelColor=444c56&amp;color=285aff" alt="GitHub stars"></a>
  <a href="LICENSE"><img src="https://img.shields.io/badge/License-Apache%202.0-285AFF?style=flat&amp;labelColor=444c56" alt="License: Apache 2.0"></a>
  <a href="pyproject.toml"><img src="https://img.shields.io/badge/Version-0.3.1-285AFF?style=flat&amp;labelColor=444c56" alt="Version 0.3.1"></a>
  <a href="docs/reports/uhp-conformance-core-2026-09-19.json"><img src="https://img.shields.io/badge/UHP-Core-16824B?style=flat&amp;labelColor=444c56" alt="UHP conformance: Core"></a>
  <a href="pyproject.toml"><img src="https://img.shields.io/badge/Python-3.10%2B-285AFF?style=flat&amp;labelColor=444c56" alt="Python 3.10 or newer"></a>
</p>

**Turn a System One decision model into an agent loop.** System One Harness observes an environment, compiles its finite action space into typed questions, gates each decision by confidence, executes the chosen action, and records the complete trace.

One model call per step. No generated actions. A probability on every transition.

<p align="center">
  <img src=".github/images/system-one-browser-gameplay-demo.gif" width="100%" alt="Jev uses System One Harness to play a live browser game by choosing one typed action per step.">
  <br>
  <sub><strong>Jev playing a live browser game through System One Harness</strong> — one typed decision per step, with no generated control text.</sub>
</p>

The first supported model is [Jev](https://typesafe.ai) by TypeSafe, available throu

Also filed under Tools & apps

0616

Jevstiller: distill Jev calls into a local model

A self-hosted proxy that trains a local model on Jev's answers. It served 72% of Banking77 locally at 15 ms, with a 2% audit and a bound on disagreement.

tomerglick57 · Tools & apps · 15 ms local

0611

Scruple: code checks for what linters miss

Semantic-aware linting with System One models: rules you can state in a sentence, checked against your code by Jev.

@nalexpeae · Tools & apps

0606

Algolia-style Postgres search with a Jev reranker

Typo tolerance, typeahead, barcodes and facet counts in plain Postgres SQL, plus an optional Jev step that asks one question per top-10 result.

@alex_forman1 · Tools & apps

0604

Virtual Closet: outfits from closet videos

built a virtual closet & outfit generator from raw closet videos. classification + semantic search thanks to jev 🥰. clueless closet dreams finally coming to life

@maitreemervana · Tools & apps