super-jev

A small, extensible decision-to-action harness for TypeSafe Jev

How it uses Jev

One judge (Jev) between your agent and your data: the agent asks in its own words, Super Jev finds the file, checks the claim, permits the action, and remembers what you approved. Your agent stays responsible for the answer.

Getting started

People: docs/GETTING-STARTED.md — the full operating manual. Agents: AGENTS.md — the numbered path and daily loop, plus wire-into-claude-code for the retrieval rule card and Stop-hook claim gate. At a terminal: npm run jev opens the app (docs/jev-cli.md). No key yet? npm run demo runs the loop once offline.

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

Keep exploring

Similar to super-jev

Browse all
Jev plays Mario project preview
Games & simulations
Less talking. More jumping. Watch Jev navigate Super Mario Bros. from game state.
advocaat project preview
SDKs & clients
A small, type-safe client for asking AI questions about your data, powered by TypeSafe Jev.
jevper project preview
SDKs & clients
Jev-shaped (TypeSafe System One) classification wrapper over OpenAI-like clients: probabilities and confidence instead of prose
jev-harness-router project preview
SDKs & clients
Per-turn router for agent harnesses: one 350ms Jev call picks the model tier, effort, tools and skill, behind a hard deadline with a regex fallback. Claude Agent SDK adapter included.
mario-jev project preview
Games & simulations
Python prototype that plays NES Super Mario Bros. from structured RAM observations, with Jev answering focused movement and jump questions.
tsai-civ2 project preview
Games & simulations
TypeSafe Jev plays original Civilization II in a browser, with live action probabilities. Experimental full-game harness.
jev-cli (tumf) project preview
Applications
Small dependency-free CLI for TypeSafe Jev
dohnuts project preview
Applications
Dohnuts builds small multimodal models for direct decisions. -> System One model
reflex project preview
Benchmarks & research
A small open decision model: state + typed questions -> calibrated probabilities. A Jev / System One re-creation on Qwen3.5.