用 TypeSafe Jev 给 AI coding agent 做持续代码质量 review 的校准审计层。持久化 19 维质量分数，输出趋势矩阵、假阳性置信检测、rubric 漂移与 ECE 校准曲线。Node 22+，零运行时依赖，CLI + GitHub Action，直接消费 jev_review 的 MCP 评估结果。
Agent tooling — page 23
Frameworks, routers and utilities that use Jev to choose tools or branches inside a larger agent system.
1,206 active projects in this collection. All collections →
Jev-powered support ticket router — parallel structured decisions via TypeSafe AI
Claude Code & Codex skills powered by Jev, TypeSafe's System One model. 256 calibrated judgements for $0.0005 in 0.72s — 360x cheaper than Claude Opus 5. Includes the first published Jev calibration curve, measured on 4,995 real agent decisions.
JEV Decision is a live demo that turns market data into structured decisions. It pulls real-time prices, valuation ratios and sector context, then runs a 20-question against TypeSafe Jev model to score buy/sell conviction, financial health and risk
Millisecond-class test-failure triage for coding agents: RETRY / FIX_CODE / FIX_ENV, powered by TypeSafe Jev.
Documentation and Codex skills for code-controlled Jev workflow patterns
通用客服平台:ChatKit UI + Jev 决策 + Qwen 生成 + RAG 知识库 + 插件化 Tools + MCP 数据面,换行业只改 YAML 不改代码
1
CLI that filters logs and text by meaning using plain-English yes/no questions and Jev probabilities.
1
Jev decision visualisation and debugging panel for coding agents. MCP server + local web timeline, confidence alerts and trace export.
1
lgtm-dsh is a DSH plugin that automatically installs lgtm, reuses the Jev already configured in DSH, and lets the agent automatically call lgtm during the testing phase.
1
Ship the result, not the conversation. 35ms discriminative decision engine for AI agent deliverables (inspired by TypeSafe Jev & Occam's razor).
Route Oh My Pi prompts between simple and advanced models with TypeSafe AI's Jev classifier.
JEV x APM progressive-context PoC: TypeSafe System One progressively materializes/dematerializes APM rules and skills
1
Persistent governance for AI agents — turn Jev judgments into replayable relation state with R2R.
An autonomous risk-monitoring MCP server. Define a risk profile — a topic, the geographic locations to watch, and the policy triggers that matter — and the server runs an agentic sweep: live web search scoped by a judgment model (Jev), source scoring, and a Markdown briefing you can read from any MCP client.
Route an agent's intent to the one installed skill that fits, using TypeSafe's Jev
Tool-agnostic System One decision orchestration skill for AI coding agents, with optional TypeSafe, Jev MCP and CodeGraph integrations.
Research and scaffolding for a TypeSafe Jev MCP server
Turn messy state into typed, auditable classifications, scores, and policy decisions over REST and MCP.
⚡ AgenticJev · 拾意 — 推荐智能体实验室。搜索 × 推荐 × 广告，Jev / MiniCPM / GPT / Claude 多模型决策与生成式组合对照。
1
Ultra-fast, fail-open advisory decisions and verbatim extractive reading view for AI coding agents and CLI pipelines
Bounded semantic routing with Jev and Poise. Rust, single-server, alpha.
Cursor plugin that routes subagents with TypeSafe Jev and exposes Choice, Score, and Noul as MCP tools.
Find where TypeSafe Jev fits your real workflows. An evidence-based audit skill for Claude Code, Codex, and compatible agents.