jevbooks

← All projects

r2r-jev

Thneoly/r2r-jev

Persistent governance for AI agents: Jev judgments become replayable relation state.

What it is

r2r-jev is a small Rust integration that turns Jev probabilistic judgments into persistent, replayable governance state for AI agents. It is for developers studying agent governance, evidence admission, and relation-based authorization.

How it uses Jev

Jev judges agent state with two Noul checks per tool call: beyond_scope and destructive. Each judgment becomes typed evidence candidates. Evidence Admission v0.1 accepts, holds, or rejects them; only accepted evidence may change R2R relations (Trust, Delegation, Authorization) that determine enforcement decisions.

Primitives:noul

Technique worth stealing

Separate probabilistic judgment from authority: admit evidence before it can mutate persistent governance relations.

Try it

cargo run -- fixture

View on GitHub

judged by Jevjev-1.13.0

Evidence

Each line is one question put to Jev about the README. ≥ 0.60 reads as yes, ≤ 0.40 as no; in between Jev is not making a call.

  • Jev-centricyes0.91
  • Shows a System One patternyes0.81
  • Handles uncertaintyno0.26
  • Measuredno0.05
  • Runnableyes0.70
  • Worth recommendingunclear0.57
  • Model replicano0.07
  • Problem scopescore on a 0–2 scale1.68
  • About Jevyes0.97

Signals by Jev jev-1.13.0, card written by DeepSeek V4.1 Flash from the README on 23 Sept 2026.