JevHarness

LLM-authored task-specific Jev harnesses with optional full-trajectory reward reflection and GEPA evolution.

How it uses Jev

Let an LLM write a task-specific harness for Jev. Run it, inspect its decisions, and optionally improve it using rewards and complete execution traces.

Source: community catalog. Details and metrics reflect the published community snapshot and are not a performance endorsement. Documentation details extracted from GitHub · the README · the project website.

Languages: Python 86% · HTML 10% · JavaScript 4%
Last activity: Oct 2026
Added: 2026-09-21
Stars at snapshot: 191
Status: Community listing

Is this your project?

Add the badge to your README to link back here.

[![Listed in Jev Library](https://jevlibrary.dev/api/badge/tianyucodings-jevharness)](https://jevlibrary.dev/projects/tianyucodings-jevharness)

Something wrong, or rather not be listed? Tell us.

Keep exploring

Similar to JevHarness

Browse all

jev-usecases

Agent tooling

Production TypeSafe Jev (System One) use-case harnesses with confidence-gated decision logic

KMade bykenhuangus5

jev-model-router (lucianfialho)

Agent tooling

Cost-optimized OpenRouter model router using TypeSafe's Jev, with a live full-catalog scorer instead of a hardcoded model list

LMade bylucianfialho2

Augustus

Agent tooling

Independent agent skill for finding, building, evaluating, and improving decision-model systems with composition rules, evaluation harnesses, and bounded prompt/program optimization; TypeSafe Jev is the default hosted exemplar.

2Made by246019

jevtree

Applications

LLM-authored decision SOP generator for Jev System One model

QMade byqzqdz2

jev-align (Sutro)

Applications

Active-learning CLI that uses human labels and GEPA to improve Jev decision definitions

SMade bysutro-sh284

jev-reward-model-evaluation

Benchmarks & research

Jev 1.13 reward-model evaluation across 8 benchmark tracks, with an interactive report and 54-row SOTA comparison

GMade bygoya41404

Loki

Integrations

Self-improving agent harness with an optional TypeSafe Jev companion for typed Choice, Score, and Noul judgments.

WMade bywundercorp26

jev_project_context

Benchmarks & research

Evidence-first long-term experiment memory skill for AI coding agents, with optional Jev decision-model layers

PMade bypoiuyjie9

tsai-civ2

Games & simulations

TypeSafe Jev plays original Civilization II in a browser, with live action probabilities. Experimental full-game harness.

PMade byphyous2