Agent Tooling for Jev — page 31

Fast 'System One' reflex for reasoning LLMs: typed probabilistic second opinions from Jev via OpenCode Zen, with PII-minimizing state redaction.

Event-driven trajectory watchdog for coding agents: per-event LLM judgments (Jev, Claude, GPT) accumulated into quarantine decisions

Calibrated decision layer and risk gate for AI coding agents. Routes yes/no, routing and scoring judgments to the Jev System One model at ~$0.0000123 per decision, then enforces the result deterministically through Claude Code PreToolUse hooks. Fitted thresholds, not guessed.

Official Model Context Protocol (MCP) server for JevGuard and TypeSafe AI

Experimental System One implementation of the SingGuard-NSFA agent-security taxonom.

把判断和写作分开的本地 agent：Jev 负责所有能用类型表达的判断，语言模型只写该写的。零第三方依赖。A local agent harness that puts a System One model in front of the LLM.

Jev-style Choice, Score and Noul decisions from ordinary models. An optional DSH plugin.

A malicious instruction protection skill and prehook for zcode

Typed client, CLI and MCP server for TypeSafe's Jev (System One) judgment model

Local Kev decision model for Claude Code, Codex, and Google Antigravity: Docker, MCP tools, shared skills, and a web playground.

Goal-driven UI verification on iOS/Android/Chromium: Jev (TypeSafe) decides each step, Maestro executes. Installs as a pi / Claude Code / Codex skill.

Forward an email, get a calibrated phishing verdict. Jev (TypeSafe System One) decides, an LLM explains. MCP server for Copilot Studio + REST for Power Automate.

Local model and reasoning router for OpenAI Codex, powered by TypeSafe Jev.

Reusable Multica agent capacity scheduling Skill with GPT-only capacity retries, Jev classification, failure caching, rerouting, and automatic recovery.

Central MCP facade for Last30Days, ScrapeCreators, Bright Data, Apify and OpenRouter TypeSafe Jev

OpenClaw plugin: TypeSafe AI Jev System One via OpenRouter decisions endpoint (fork of jason-allen-oneal/openclaw-plugin-typesafe-ai)

Privacy-aware TypeSafe Jev tools and per-turn tool-family routing for OpenClaw.

Dynamic two-stage skill & tool pre-filtering for OpenClaw using Decision Models (Jev / TypeSafe)

Unofficial OpenCode v2 plugin: LangSearch web search provider, optionally gated by Jev (TypeSafe AI) to cut tokens and filter noisy/injected results before they reach the model.

MCP server for JEV (TypeSafe System One) classification via OpenRouter

Community Paperclip plugin for Jev-powered issue triage and automatic routing.

Dynamic context filter for Pi Coder using Jev by TypeSafe to save 75%+ on context tokens.

RCIL: The Question Intelligence Layer for System One (TypeSafe Jev). Cuts agent costs by 95% and eliminates decision hallucinations.

Say it, and your computer does it. Voice-controlled computer agent on TypeSafe Jev for macOS and Windows.