rh-guard

Reward-hack radar for coding agents: structural denies + TypeSafe Jev System One sidecar for Claude Code & Cursor hooks

How it uses Jev

Companion, not a merge. Augustus is design-judgment for where typed System One judgment belongs. This repo is the live hazard gate on agent tools. Keep them separate.

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: TypeScript 99% · CSS 1% · Shell 1% · JavaScript 1%
License: MIT
Last activity: Oct 2026
Added: 2026-09-20
Stars at snapshot: 3
Status: Community listing

agent-hooksagent-skillsai-agentsai-safetyclaude-codeclaude-code-hooks

Is this your project?

Add the badge to your README to link back here.

[![Listed in Jev Library](https://jevlibrary.dev/api/badge/24601-rh-guard)](https://jevlibrary.dev/projects/24601-rh-guard)

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

Keep exploring

Similar to rh-guard

Browse all

jev-flash-router

Benchmarks & research

open-sourced jev-flash-router: an MCP server for TypeSafe's new Jev model. AI coding agents waste hundreds of reasoning tokens just deciding which file to edit, which route to pick, or whether a diff breaks tests. Jev evaluates state and outputs calibrated probabilities. Works with Cursor, Windsurf, & Claude Code

RMade byRavinder823

toolgate (RiskAverseTech)

Benchmarks & research

Open auto mode for AI agents — a calibrated tool-call firewall powered by TypeSafe Jev. Ships as a Claude Code hook

RMade byRiskAverseTech2

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

jev-mcp (burnigtm)

Agent tooling

MCP server that puts TypeSafe Jev on the coding loop in Cursor, Codex, and any MCP client

BMade byburnigtm43

agent-fastpath

Browser & computer use

Jev MCP server: a decision layer for coding agents, built on TypeSafe Jev (System One model). Ship gates, risk checks, file triage that keeps files out of context, and a safe headless browser, with calibrated confidence. For Claude Code, Codex, Cursor.

AMade byabhishekswe3

jev-guard

Agent tooling

Prompt-injection and dangerous-action guard for coding agents (Claude Code, Codex, pi, ACP), powered by Jev

LMade byleepokai32

construct-auto-classifier

Agent tooling

Effect-based safety gate for AI coding agents' shell commands (OpenCode, Antigravity): fast structural rules, then TypeSafe's Jev or a chat model judges what a command does. Certified with Jev at zero dangerous commands allowed.

GMade bygodspede3

AskJev

Browser & computer use

AskJev — Jev autopilot for any website + guard on irreversible clicks (TypeSafe System One, not Claude)

RMade byranjan28298

siege

Games & simulations

SIEGE: 200 people vs one agent. A typed action gate (TypeSafe System One) that learns from every breach, evaluated by W&B Weave, hardened by a defender loop. Built at CoreWeave Hacks: Agent Loops 2026.

VMade byvnmoorthy0