Skip to content

R2r Jev

Agent governance integration where Jev judges two Noul checks per tool call (beyond scope, destructive), an admission policy decides which judgments may become evidence, and admitted evidence degrades Trust, Delegation and Authorization in persistent R2R relation state.

Stars
1
Language
Rust
License
APACHE-2.0
Published by
Community
Last verified
Oct 2026

About this project

r2r-jev is a small public integration built around one rule: Jev turns unstructured state into typed probabilistic decisions. Evidence Admission decides which observations are eligible to enter governance. R2R turns admitted evidence and events into persistent, replayable relation state. The primary execution path is now: Most agent guardrails answer a local question: Evidence Admission adds a second question before persistent state is allowed to change:

Quoted from the project's README. Read the full README.

Highlights

  • Runs the full governance path offline with a recorded Jev-style fixture; live mode adds a TypeSafe key and CLI flags
  • Admission v0.1 returns Reject, Hold or Accept (Strong or Corroborated), and only Accept may emit governance-active EvidenceAdmitted
  • Built-in counterexamples: 99.9% confidence from a low-reliability source is rejected, and 76% confidence needs two corroborators to be accepted
  • Evidence confidence is converted to integer parts-per-million at the adapter boundary; ids and virtual ticks are deterministic
  • Three-act demo shows DENY from accepted evidence, unchanged state from rejected evidence, and a human override that restores authorization under supervision

Watch out

Apache-2.0. Intentionally small: it demonstrates the integration boundary rather than the full R2R runtime, and Admission v0.1 is described as deliberately minimal.

Questions about this entry

Short answers drawn from the linked project or page.

More Jev resources like this

A Kubernetes admission webhook that asks Jev to allow, flag, or deny a manifest change for intent-shaped policies a Rego rule cannot express, keeping hard rules in Rego and failing closed by default.
Jev Integrations & Adapters#community#go#kubernetes
Community
20.5kGitHub stars
Python agent framework whose TypeSafeModel class runs Jev as a decision model inside Pydantic AI agents: Pydantic output-type fields become typed Jev questions answered together in one request, each answer carrying its own confidence, and tools can be offered as Choice options.
Jev Integrations & Adapters#community#agents#framework
159GitHub stars
Rust agent workbench for ongoing work that keeps branch decisions in their own crate and backend notes in the adapter. Its first decision backend is Cloudflare's Clef, borrowing Clef's System One-shaped state and questions request instead of reinventing the interface.
Jev Integrations & Adapters#community#agent-runtime#rust
27GitHub stars
A DeepSeek Harness plugin where Jev Choice and Noul checks gate every non-read-only tool call - deny confident high-risk calls, escalate ambiguous ones, auto-approve clearly granted reversible ones - and optionally score session messages so valuable context survives instead of being dropped oldest-first.
Jev Tools & CLIs#community#plugin#guardrails
22GitHub stars
An MCP proxy and hooks adapter that screens an agent's tool calls before they run and the tool results before the agent reads them, with thresholds in a policy file.
Jev Agents & MCP#community#mcp#security
10GitHub stars
x402 risk-check provider that scores agent counterparties before payment: OFAC, phishing and drainer feeds, transaction simulation, drainer-kit fingerprints and a Jev model reading context for injected instructions, issuing an ES256-signed attestation per verdict.
Jev Integrations & Adapters#community#x402#payments
Back to all resources

From the Jev community

Posts from builders shipping with Jev right now.

Follow @typesafeai