A gallery of open-source projects that use Jev, the TypeSafe AI System One model: typed questions in, calibrated probabilities out, in about 300 ms.https://jevbooks.comenhttps://jevbooks.com/en/p/tpellet__jevifyhttps://jevbooks.com/en/p/tpellet__jevifygrep for meaning: find errors, commits, commands by description.Mon, 21 Sep 2026 03:34:16 GMT<p>grep for meaning: find errors, commits, commands by description.</p><p>jevify is a Rust CLI that searches by meaning: it finds the error in a long log, a commit by description, or the command for a task. It works like a Unix tool, reading stdin and printing lines.</p><p><a href="https://github.com/tpellet/jevify">tpellet/jevify</a></p>Scoring & Rankinghttps://jevbooks.com/en/p/tickerdev__jevfanityhttps://jevbooks.com/en/p/tickerdev__jevfanityProfanity and moderation API powered by Jev on Cloudflare Workers.Mon, 21 Sep 2026 03:34:14 GMT<p>Profanity and moderation API powered by Jev on Cloudflare Workers.</p><p>Jevfanity is a small profanity and moderation API for developers who need to flag text. It runs on Cloudflare Workers, exposes a hosted endpoint, and ships a typed npm SDK.</p><p><a href="https://github.com/TickerDev/jevfanity">TickerDev/jevfanity</a></p>Content Moderationflat-fanouttiered-thresholdshttps://jevbooks.com/en/p/mblode__taste-linthttps://jevbooks.com/en/p/mblode__taste-lintCatch AI slop in UI, copy, and agent instructions before ship.Mon, 21 Sep 2026 03:32:40 GMT<p>Catch AI slop in UI, copy, and agent instructions before ship.</p><p>A TypeScript CLI that runs design review on every commit, checking product interfaces, whole pages, and text for AI slop. Built for developers and teams who want measurable rules enforced locally and judgement calls scored.</p><p><a href="https://github.com/mblode/taste-lint">mblode/taste-lint</a></p>Evaluation & Benchmarkinghttps://jevbooks.com/en/p/eugeniughelbur__jev-engineeringhttps://jevbooks.com/en/p/eugeniughelbur__jev-engineeringTool-call safety gate for AI coding agents, backed by Jev.Mon, 21 Sep 2026 03:32:40 GMT<p>Tool-call safety gate for AI coding agents, backed by Jev.</p><p>jev-gate is a safety gate for AI coding agents like Claude Code, Codex and Cursor. It answers allow, ask or deny for every tool call, and ships a 300-call injection test showing what the gate catches and what walks past it.</p><p><a href="https://github.com/eugeniughelbur/jev-engineering">eugeniughelbur/jev-engineering</a></p>Verification & Guardrailsflat-fanouttiered-thresholdsfailure-mode-knobhttps://jevbooks.com/en/p/keltokhy__jgrephttps://jevbooks.com/en/p/keltokhy__jgrepgrep where the pattern is a plain-English description.Mon, 21 Sep 2026 03:32:40 GMT<p>grep where the pattern is a plain-English description.</p><p>jgrep filters text lines, structured records, functions, and diff hunks against plain-English descriptions. It is for developers and analysts who want semantic filtering in a pipe, on logs, files, CSVs, JSONL, patches, or code.</p><p><a href="https://github.com/keltokhy/jgrep">keltokhy/jgrep</a></p>Data Labeling & Curationflat-fanouthttps://jevbooks.com/en/p/jerryjliu__docjevhttps://jevbooks.com/en/p/jerryjliu__docjevDocument classification and splitting with Jev, LiteParse, and optional LlamaParse.Mon, 21 Sep 2026 03:32:40 GMT<p>Document classification and splitting with Jev, LiteParse, and optional LlamaParse.</p><p>A Python library, CLI, and local app that classifies a PDF, DOCX, or PPTX against natural-language category rules, or splits a packet into ordered sub-documents with page ranges. Built for document-pipeline developers.</p><p><a href="https://github.com/jerryjliu/docjev">jerryjliu/docjev</a></p>Classification & Routinghttps://jevbooks.com/en/p/2045max__flappy-jevhttps://jevbooks.com/en/p/2045max__flappy-jevFlappy Bird played by Jev: one yes/no question per frame.Sun, 20 Sep 2026 20:55:07 GMT<p>Flappy Bird played by Jev: one yes/no question per frame.</p><p>A browser Flappy Bird clone whose flapping decisions are made by TypeSafe's Jev model instead of a human player. Deployed as a Cloudflare Worker; game mechanics and assets come from CodeExplainedRepo/FlappyBird-JavaScript.</p><p><a href="https://github.com/2045max/flappy-jev">2045max/flappy-jev</a></p>Game & Simulationflat-fanouthttps://jevbooks.com/en/p/trilokx__omni-scraper-mcphttps://jevbooks.com/en/p/trilokx__omni-scraper-mcpOne MCP facade for multi-source scraping, normalization and Jev evaluation.Sun, 20 Sep 2026 20:55:06 GMT<p>One MCP facade for multi-source scraping, normalization and Jev evaluation.</p><p>Omni Scraper is an integration and orchestration layer that gives agents one local MCP server to collect public and explicitly authorized data, normalize records into a NormalizedDocument schema, run the vendored Last30Days research engine, and discover upstream provider tools dynamically.</p><p><a href="https://github.com/Trilokx/omni-scraper-mcp">Trilokx/omni-scraper-mcp</a></p>Evaluation & Benchmarkinghttps://jevbooks.com/en/p/philippdubach__pi-jev-routerhttps://jevbooks.com/en/p/philippdubach__pi-jev-routerPareto-optimal OpenRouter model router for pi, based on Jev.Sun, 20 Sep 2026 20:55:05 GMT<p>Pareto-optimal OpenRouter model router for pi, based on Jev.</p><p>A minimal router for pi that ranks the whole OpenRouter catalog per task, filters to fitting models, computes a Pareto frontier over quality, cost and latency, then picks the knee point. For pi users who want automatic model selection.</p><p><a href="https://github.com/philippdubach/pi-jev-router">philippdubach/pi-jev-router</a></p>Classification & Routinghttps://jevbooks.com/en/p/bcivitcioglu__jev-chesshttps://jevbooks.com/en/p/bcivitcioglu__jev-chessTwo AI models play chess: one picks, one writes.Sun, 20 Sep 2026 20:55:03 GMT<p>Two AI models play chess: one picks, one writes.</p><p>A chess match between Jev, which selects from legal moves, and a text-writing opponent (Qwen, GPT or Claude). Every game, move and rejected attempt is recorded to SQLite, with 15-second video export.</p><p><a href="https://github.com/bcivitcioglu/jev-chess">bcivitcioglu/jev-chess</a></p>Game & Simulationhttps://jevbooks.com/en/p/ddlaws0n__jevportfoliohttps://jevbooks.com/en/p/ddlaws0n__jevportfolio1,000 synthetic SaaS accounts triaged by Jev's six constrained judgments.Sun, 20 Sep 2026 20:55:03 GMT<p>1,000 synthetic SaaS accounts triaged by Jev's six constrained judgments.</p><p>A portfolio triage demo: 1,000 synthetic SaaS accounts go in, six constrained Jev judgments per account come back, and ordinary TypeScript decides which accounts a human should open today.</p><p><a href="https://github.com/ddlaws0n/jevportfolio">ddlaws0n/jevportfolio</a></p>Classification & Routingflat-fanouttiered-thresholdscode-owned-policyhttps://jevbooks.com/en/p/valsecchi75__squinthttps://jevbooks.com/en/p/valsecchi75__squintA PreToolUse hook that narrows Claude's file reads to the answering chunk.Sun, 20 Sep 2026 20:55:02 GMT<p>A PreToolUse hook that narrows Claude's file reads to the answering chunk.</p><p>squint is a Claude Code PreToolUse hook for developers working in large source files. It intercepts Read calls, asks a small fast model which chunk answers the question, and rewrites the tool's offset and limit before the file is opened.</p><p><a href="https://github.com/valsecchi75/squint">valsecchi75/squint</a></p>Agent Decisionshttps://jevbooks.com/en/p/yasdelayu__jev-crypto-scouthttps://jevbooks.com/en/p/yasdelayu__jev-crypto-scoutCrypto screening: quant signals in code, news judgment via Jev.Sun, 20 Sep 2026 20:55:02 GMT<p>Crypto screening: quant signals in code, news judgment via Jev.</p><p>A Python tool for crypto traders and analysts that screens top coins by market cap using quantitative signals from CoinGecko and Bybit, and classifies news sentiment/catalyst/confirmation via Jev. It prints a table for human reading, not a trading bot.</p><p><a href="https://github.com/yasdelayu/jev-crypto-scout">yasdelayu/jev-crypto-scout</a></p>Scoring & Rankingflat-fanouthttps://jevbooks.com/en/p/sharkelot__jev-laya-freehttps://jevbooks.com/en/p/sharkelot__jev-laya-freeFree local Jev-compatible typed decisions with rules or Laya.Sun, 20 Sep 2026 20:55:00 GMT<p>Free local Jev-compatible typed decisions with rules or Laya.</p><p>An independent, MIT-licensed local implementation of the documented Jev typed-decision HTTP shape, with a Python client and optional Laya inference. For developers testing typed decisions locally without a hosted API.</p><p><a href="https://github.com/Sharkelot/jev-laya-free">Sharkelot/jev-laya-free</a></p>Agent Decisionsflat-fanouthttps://jevbooks.com/en/p/jmanhype__jev-dspy-labhttps://jevbooks.com/en/p/jmanhype__jev-dspy-labCalibration and selective-risk benchmarks for Jev decisions in DSPy workflows.Sun, 20 Sep 2026 20:54:59 GMT<p>Calibration and selective-risk benchmarks for Jev decisions in DSPy workflows.</p><p>A companion measurement lab for dspy-typesafeify and other callers that use TypeSafe decisions inside DSPy-style pipelines. It provides reproducible calibration, confidence-gating, latency, and modeled-cost benchmarks, plus offline replay.</p><p><a href="https://github.com/jmanhype/jev-dspy-lab">jmanhype/jev-dspy-lab</a></p>Calibration & Researchhttps://jevbooks.com/en/p/mrribvar__hermes-jev-pluginhttps://jevbooks.com/en/p/mrribvar__hermes-jev-pluginHermes Agent plugin exposing TypeSafe AI Jev for structured evaluation.Sun, 20 Sep 2026 20:54:59 GMT<p>Hermes Agent plugin exposing TypeSafe AI Jev for structured evaluation.</p><p>A standalone Hermes Agent plugin that registers a jev_evaluate tool, letting the agent classify, route, verify, score, and gate using TypeSafe AI Jev via the Vercel AI Gateway.</p><p><a href="https://github.com/Mrribvar/hermes-jev-plugin">Mrribvar/hermes-jev-plugin</a></p>Evaluation & Benchmarkingflat-fanouthttps://jevbooks.com/en/p/priorbench__jevhttps://jevbooks.com/en/p/priorbench__jevIndependent, pre-registered evaluation of TypeSafe AI's Jev model.Sun, 20 Sep 2026 20:54:58 GMT<p>Independent, pre-registered evaluation of TypeSafe AI's Jev model.</p><p>PriorBench is an independent, pre-registered evaluation of TypeSafe AI's Jev, a System One model. It runs 21 experiments and 5,721 calls, publishing raw data, predictions, and a report for researchers and practitioners.</p><p><a href="https://github.com/priorbench/jev">priorbench/jev</a></p>Evaluation & Benchmarkinghttps://jevbooks.com/en/p/rubinagentagi-tech__jev-heart-risk-benchhttps://jevbooks.com/en/p/rubinagentagi-tech__jev-heart-risk-benchBenchmarking Jev on 5,000 CDC heart-risk survey respondents.Sun, 20 Sep 2026 20:54:56 GMT<p>Benchmarking Jev on 5,000 CDC heart-risk survey respondents.</p><p>Scores Jev (TypeSafe System One) on 5,000 real CDC BRFSS 2015 respondents against logistic regression, a hand-written rule, a chat LLM, and the base rate. For researchers evaluating decision models.</p><p><a href="https://github.com/rubinagentagi-tech/jev-heart-risk-bench">rubinagentagi-tech/jev-heart-risk-bench</a></p>Evaluation & Benchmarkingmeasured-calibrationhttps://jevbooks.com/en/p/daneknudsen8-maker__jev-voice-controlhttps://jevbooks.com/en/p/daneknudsen8-maker__jev-voice-controlVoice control for your browser using TypeSafe's Jev model.Sun, 20 Sep 2026 20:54:56 GMT<p>Voice control for your browser using TypeSafe's Jev model.</p><p>A Chrome extension that lets you control your browser by voice. It captures speech, sends the transcript and the page's interactive elements to Jev, and executes the returned typed command.</p><p><a href="https://github.com/daneknudsen8-maker/jev-voice-control">daneknudsen8-maker/jev-voice-control</a></p>Agent Decisionscandidate-enumerationhttps://jevbooks.com/en/p/sathariels__jevcheckhttps://jevbooks.com/en/p/sathariels__jevcheckBehavioral contracts for TypeSafe Jev: pin, eval, catch flips.Sun, 20 Sep 2026 20:54:56 GMT<p>Behavioral contracts for TypeSafe Jev: pin, eval, catch flips.</p><p>jevcheck is a pytest-like tool for TypeSafe Jev. It pins a production contract (baseline model, fixtures, expected answers), evals a candidate model, and fails when answers flip or confidence drops.</p><p><a href="https://github.com/sathariels/jevcheck">sathariels/jevcheck</a></p>Evaluation & Benchmarkinghttps://jevbooks.com/en/p/jexp__watfilehttps://jevbooks.com/en/p/jexp__watfileClassify files with a decision-making AI model and sort them into category folders.Sun, 20 Sep 2026 20:54:55 GMT<p>Classify files with a decision-making AI model and sort them into category folders.</p><p>watfile is a Python CLI for people who want to sort documents into category folders. It extracts text from text files and PDFs, asks a decision model for a category, and places files by symlink, copy, or move.</p><p><a href="https://github.com/jexp/watfile">jexp/watfile</a></p>Classification & Routinghttps://jevbooks.com/en/p/anasbekheit__typesafe-jev-mcphttps://jevbooks.com/en/p/anasbekheit__typesafe-jev-mcpMCP server exposing TypeSafe Jev as a typed evaluate tool.Sun, 20 Sep 2026 20:54:53 GMT<p>MCP server exposing TypeSafe Jev as a typed evaluate tool.</p><p>An MCP server for TypeSafe Jev, written in Rust. It gives agents one tool, evaluate, that takes a state plus typed questions and returns noul, choice, or score answers with probabilities. Not affiliated with TypeSafe.</p><p><a href="https://github.com/anasbekheit/typesafe-jev-mcp">anasbekheit/typesafe-jev-mcp</a></p>Evaluation & Benchmarkingflat-fanouthttps://jevbooks.com/en/p/villagertim__neonhop-3dhttps://jevbooks.com/en/p/villagertim__neonhop-3dNeonHop 3D: testbed for Jev-only runtime inference in a game.Sun, 20 Sep 2026 20:54:53 GMT<p>NeonHop 3D: testbed for Jev-only runtime inference in a game.</p><p>A browser-based 3D game used as an empirical research testbed to evaluate whether Jev, combined with local application logic, can be the sole runtime inference model for decision-making.</p><p><a href="https://github.com/villagertim/neonhop-3d">villagertim/neonhop-3d</a></p>Evaluation & Benchmarkinghttps://jevbooks.com/en/p/g0runmezadam__what-is-jevhttps://jevbooks.com/en/p/g0runmezadam__what-is-jevIndependent, source-linked research on TypeSafe AI's Jev and its ecosystem.Sun, 20 Sep 2026 20:54:52 GMT<p>Independent, source-linked research on TypeSafe AI's Jev and its ecosystem.</p><p>A sourced, critical research file on Jev, TypeSafe AI's System One decision model, plus a rubric-scored map of 947 public GitHub repositories that use it, with recurring patterns, datasets, and bilingual documentation.</p><p><a href="https://github.com/g0runmezadam/what-is-jev">g0runmezadam/what-is-jev</a></p>Calibration & Researchhttps://jevbooks.com/en/p/abhishekmishragithub__semantic-microscopehttps://jevbooks.com/en/p/abhishekmishragithub__semantic-microscopeLabel every sentence with calibrated probabilities from Jev, rendered as a heatmap.Sun, 20 Sep 2026 20:54:51 GMT<p>Label every sentence with calibrated probabilities from Jev, rendered as a heatmap.</p><p>A Python tool that labels every sentence of a document with calibrated probabilities from Jev (TypeSafe AI's System One model) and renders the results as a heatmap for whole-document viewing.</p><p><a href="https://github.com/abhishekmishragithub/semantic-microscope">abhishekmishragithub/semantic-microscope</a></p>Data Labeling & Curationflat-fanouthttps://jevbooks.com/en/p/octanevz__jev-playground-openrouterhttps://jevbooks.com/en/p/octanevz__jev-playground-openrouterLocal browser playground for Jev decision model via OpenRouter.Sun, 20 Sep 2026 20:54:49 GMT<p>Local browser playground for Jev decision model via OpenRouter.</p><p>A local browser playground for TypeSafe's Jev decision model, accessed through OpenRouter. It is for developers who want to experiment with Jev's typed questions without waiting for a TypeSafe invitation. Python standard library only.</p><p><a href="https://github.com/octanevz/jev-playground-openrouter">octanevz/jev-playground-openrouter</a></p>Otherhttps://jevbooks.com/en/p/thewaywithin__jev-benchhttps://jevbooks.com/en/p/thewaywithin__jev-benchDoes the cited source actually say it? A 42-claim benchmark.Sun, 20 Sep 2026 20:54:49 GMT<p>Does the cited source actually say it? A 42-claim benchmark.</p><p>A 42-claim benchmark testing whether a cited passage supports a sentence as written, run against Jev and three frontier models, with raw outputs and scored results.</p><p><a href="https://github.com/TheWayWithin/jev-bench">TheWayWithin/jev-bench</a></p>Evaluation & Benchmarkingtiered-thresholdshttps://jevbooks.com/en/p/pasblinn__jev-labhttps://jevbooks.com/en/p/pasblinn__jev-labOpen lab for correct Jev routing in front of Claude Code.Sun, 20 Sep 2026 20:54:49 GMT<p>Open lab for correct Jev routing in front of Claude Code.</p><p>An open lab for using Jev, TypeSafe's System One decision model, correctly next to Claude Code. It documents measured bugs, a patch, a hard fallback with alerts, and where typed decisions belong.</p><p><a href="https://github.com/Pasblinn/jev-lab">Pasblinn/jev-lab</a></p>Classification & Routinghttps://jevbooks.com/en/p/brightshore__jev-nethttps://jevbooks.com/en/p/brightshore__jev-netLightweight .NET client for TypeSafe AI's Jev probability API.Sun, 20 Sep 2026 20:54:48 GMT<p>Lightweight .NET client for TypeSafe AI's Jev probability API.</p><p>A community .NET client for the TypeSafe AI API (System One / Jev), letting C# code ask small typed questions about text or structured state and receive calibrated probabilities to act on.</p><p><a href="https://github.com/brightshore/jev-net">brightshore/jev-net</a></p>Infra / SDKs / Integrationsflat-fanoutno-judgment-bandhttps://jevbooks.com/en/p/ariasnico__jevtesthttps://jevbooks.com/en/p/ariasnico__jevtestASCII flirt game where Jev judges your pickup lines.Sun, 20 Sep 2026 20:54:47 GMT<p>ASCII flirt game where Jev judges your pickup lines.</p><p>Caramelo is an ASCII flirt game in Spanish. You're at the door of a nightclub, not on the list, with six attempts to convince the bouncer. Winning unlocks a VIP chapter parodying Clavicular. For friends, mobile-first.</p><p><a href="https://github.com/ariasnico/jevtest">ariasnico/jevtest</a></p>Game & Simulationhttps://jevbooks.com/en/p/dr-dimitru__claude-jev-pluginhttps://jevbooks.com/en/p/dr-dimitru__claude-jev-pluginTypeSafe Jev semantic guardrails for Claude Code tool execution.Sun, 20 Sep 2026 20:54:46 GMT<p>TypeSafe Jev semantic guardrails for Claude Code tool execution.</p><p>A Claude Code plugin that adds TypeSafe Jev semantic judgments to Bash, Write, and Edit tool calls, warning or asking before potentially destructive, exfiltrating, or out-of-scope actions.</p><p><a href="https://github.com/dr-dimitru/claude-jev-plugin">dr-dimitru/claude-jev-plugin</a></p>Verification & Guardrailsflat-fanouthttps://jevbooks.com/en/p/agentic-guides__decision-kernelhttps://jevbooks.com/en/p/agentic-guides__decision-kernelJudgment API: TypeSafe Jev on Cloudflare Workers for triage, urgency, spam, tone, risk.Sun, 20 Sep 2026 20:54:46 GMT<p>Judgment API: TypeSafe Jev on Cloudflare Workers for triage, urgency, spam, tone, risk.</p><p>Decision Kernel exposes TypeSafe Jev as a fast, calibrated, low-cost judgment API for small SaaS, support teams, and agent developers. It classifies text into triage, urgency, spam, tone, and risk, and provides an agent-safety gate for actions.</p><p><a href="https://github.com/Agentic-Guides/decision-kernel">Agentic-Guides/decision-kernel</a></p>Classification & Routingflat-fanouttiered-thresholdshttps://jevbooks.com/en/p/collapseindex__jev-ultralightspeedhttps://jevbooks.com/en/p/collapseindex__jev-ultralightspeedBatch typed judgements over text queues with TypeSafe's Jev.Sun, 20 Sep 2026 20:54:45 GMT<p>Batch typed judgements over text queues with TypeSafe's Jev.</p><p>A Python library that answers one question about each item in a large pile of text — tickets, reviews, logs, model completions — by packing many items into each Jev request. For queues, not single requests.</p><p><a href="https://github.com/collapseindex/jev-ultralightspeed">collapseindex/jev-ultralightspeed</a></p>Classification & Routinghttps://jevbooks.com/en/p/walidboulanouar__awesome-jev-use-caseshttps://jevbooks.com/en/p/walidboulanouar__awesome-jev-use-casesCurated list of TypeSafe AI Jev demos, repos, limits and examples.Sun, 20 Sep 2026 20:54:44 GMT<p>Curated list of TypeSafe AI Jev demos, repos, limits and examples.</p><p>An unofficial, CC0-licensed awesome list of things built with TypeSafe AI Jev, with 74 demos ranked by likes, 150+ GitHub repos, limits, cost and API examples. For developers exploring Jev use cases.</p><p><a href="https://github.com/walidboulanouar/awesome-jev-use-cases">walidboulanouar/awesome-jev-use-cases</a></p>Otherhttps://jevbooks.com/en/p/zerox-01__jev-atlashttps://jevbooks.com/en/p/zerox-01__jev-atlasA Chinese-first index of TypeSafe JEV community content.Sun, 20 Sep 2026 20:54:43 GMT<p>A Chinese-first index of TypeSafe JEV community content.</p><p>JEV Atlas is a continuously updated, Chinese-first public index of real TypeSafe JEV projects, videos, articles, and open-source demos, built with React, TypeScript, and Vite for static deployment.</p><p><a href="https://github.com/ZeroX-01/jev-atlas">ZeroX-01/jev-atlas</a></p>Otherhttps://jevbooks.com/en/p/galitianu__jev4jhttps://jevbooks.com/en/p/galitianu__jev4jJava SDK for TypeSafe AI: typed questions, typed answers.Sun, 20 Sep 2026 20:54:42 GMT<p>Java SDK for TypeSafe AI: typed questions, typed answers.</p><p>A Java 17+ SDK for the TypeSafe AI API. It sends a state (text or JSON) plus typed questions and returns calibrated probabilities for noul, choice, and score answers, with Jackson as the only dependency.</p><p><a href="https://github.com/galitianu/jev4j">galitianu/jev4j</a></p>Infra / SDKs / Integrationsflat-fanouthttps://jevbooks.com/en/p/benjiscollector__rockyhttps://jevbooks.com/en/p/benjiscollector__rockyVoice-controlled computer agent on TypeSafe Jev for macOS and Windows.Sun, 20 Sep 2026 20:54:42 GMT<p>Voice-controlled computer agent on TypeSafe Jev for macOS and Windows.</p><p>Rocky is a voice-controlled computer agent for macOS (Windows written but untested) that lets you say a command and have your computer perform it, built on TypeSafe's Jev decision model.</p><p><a href="https://github.com/BenjisCollector/rocky">BenjisCollector/rocky</a></p>Agent Decisionsflat-fanoutspeculative-headscontract-validationcandidate-enumerationhttps://jevbooks.com/en/p/eugeniughelbur__jev-gatehttps://jevbooks.com/en/p/eugeniughelbur__jev-gateSafety gate for AI coding agents, backed by TypeSafe's Jev.Sun, 20 Sep 2026 20:54:41 GMT<p>Safety gate for AI coding agents, backed by TypeSafe's Jev.</p><p>A tool-call gate for Claude Code, Codex, Cursor and other AI coding agents. It answers allow, ask or deny in about 400 milliseconds for two hundredths of a cent, so every action can be checked instead of approving everything or nothing.</p><p><a href="https://github.com/eugeniughelbur/jev-gate">eugeniughelbur/jev-gate</a></p>Verification & Guardrailsflat-fanouttiered-thresholdsfailure-mode-knobhttps://jevbooks.com/en/p/jamilxt__typesafe-ai-javahttps://jevbooks.com/en/p/jamilxt__typesafe-ai-javaCommunity Java SDK for the TypeSafe AI System One (Jev) API.Sun, 20 Sep 2026 20:54:39 GMT<p>Community Java SDK for the TypeSafe AI System One (Jev) API.</p><p>A community-maintained Java SDK for the TypeSafe AI System One (Jev) API, for JVM developers. It provides a pure-Java core, Kotlin extensions, a Spring Boot starter, and a Spring AI bridge. Not an official TypeSafe product.</p><p><a href="https://github.com/jamilxt/typesafe-ai-java">jamilxt/typesafe-ai-java</a></p>Infra / SDKs / Integrationsflat-fanouthttps://jevbooks.com/en/p/psychodissection-ship-it__aithyrion-directorhttps://jevbooks.com/en/p/psychodissection-ship-it__aithyrion-directorAI music video director with beat-synced full-motion generation.Sun, 20 Sep 2026 20:54:38 GMT<p>AI music video director with beat-synced full-motion generation.</p><p>Aithyrion Director is an integrated director tool that analyzes a song's beats, structure, and energy transitions in real time, then coordinates an AI director with video generation AI to automatically supervise and generate beat-synced full-motion AI music videos.</p><p><a href="https://github.com/psychodissection-ship-it/aithyrion-director">psychodissection-ship-it/aithyrion-director</a></p>Agent Decisionshttps://jevbooks.com/en/p/daseinlabs__open-jevhttps://jevbooks.com/en/p/daseinlabs__open-jevOne-pass option scoring with local Gemma 3 4B on Apple silicon via MLX.Sun, 20 Sep 2026 20:54:38 GMT<p>One-pass option scoring with local Gemma 3 4B on Apple silicon via MLX.</p><p>openjev scores pre-written options given a context using a local Gemma 3 4B model on Apple silicon via MLX. It provides a CLI, HTTP server, and a TypeSafe System One-compatible endpoint for ranking options.</p><p><a href="https://github.com/daseinlabs/open-jev">daseinlabs/open-jev</a></p>Scoring & Rankinghttps://jevbooks.com/en/p/fbddcz__embodied-jevhttps://jevbooks.com/en/p/fbddcz__embodied-jevMuJoCo Franka Panda workbench for embodied AI decisions with Jev and other models.Sun, 20 Sep 2026 20:54:36 GMT<p>MuJoCo Franka Panda workbench for embodied AI decisions with Jev and other models.</p><p>A browser-based MuJoCo Franka Panda workbench for embodied AI experiments. It offers three simulation tasks, supports rule baselines, local MiniCPM5-2B, TypeSafe Jev, Claude, and OpenAI-compatible APIs, and visualizes each decision step.</p><p><a href="https://github.com/FBddcz/embodied-jev">FBddcz/embodied-jev</a></p>Agent Decisionshttps://jevbooks.com/en/p/fazlerocks__jevmailhttps://jevbooks.com/en/p/fazlerocks__jevmailRead-only Gmail triage with Jev probabilities, sorted locally.Sun, 20 Sep 2026 20:54:35 GMT<p>Read-only Gmail triage with Jev probabilities, sorted locally.</p><p>Jevmail is an open-source, read-only Gmail triage app for people who want inbox zero. It sorts mail into Needs reply, Updates, Promos, Sales, and Spam with a 1–5 urgency score, storing everything in a local SQLite file.</p><p><a href="https://github.com/fazlerocks/jevmail">fazlerocks/jevmail</a></p>Classification & Routingflat-fanouthttps://jevbooks.com/en/p/can1357__jegrephttps://jevbooks.com/en/p/can1357__jegrepSemantic grep: describe it, find it, powered by Jev.Sun, 20 Sep 2026 20:54:35 GMT<p>Semantic grep: describe it, find it, powered by Jev.</p><p>jegrep is a Rust semantic grep that scores live repository files and line ranges with Jev probabilities, without an embedding index or background daemon. It is for developers and coding agents searching code by natural-language description.</p><p><a href="https://github.com/can1357/jegrep">can1357/jegrep</a></p>Scoring & Rankingfailure-mode-knobhttps://jevbooks.com/en/p/iammrduncan__typesafe-ai-benchmarkhttps://jevbooks.com/en/p/iammrduncan__typesafe-ai-benchmarkSide-by-side benchmark of Jev vs Qwen on structured judgment tasks.Sun, 20 Sep 2026 20:54:33 GMT<p>Side-by-side benchmark of Jev vs Qwen on structured judgment tasks.</p><p>An LLM gateway and benchmark comparing TypeSafe Jev with Qwen 3.8 27B on Cerebras across seven synthetic workloads, recording latency, cost, and judgment quality for developers evaluating structured-output approaches.</p><p><a href="https://github.com/iammrduncan/typesafe-ai-benchmark">iammrduncan/typesafe-ai-benchmark</a></p>Evaluation & Benchmarkingflat-fanouthttps://jevbooks.com/en/p/spoonnotfound__soupbasehttps://jevbooks.com/en/p/spoonnotfound__soupbaseJev hosts a bilingual lateral-thinking puzzle site.Sun, 20 Sep 2026 20:54:32 GMT<p>Jev hosts a bilingual lateral-thinking puzzle site.</p><p>Soupbase is a text-centric Chinese/English lateral-thinking puzzle website where Jev acts as host. It supports asking questions, step-by-step hints, submitting reconstructions, private creation, and link sharing.</p><p><a href="https://github.com/spoonnotfound/soupbase">spoonnotfound/soupbase</a></p>Game & Simulationhttps://jevbooks.com/en/p/khordoo__jev-reflex-autonomy-labhttps://jevbooks.com/en/p/khordoo__jev-reflex-autonomy-labMulti-drone lab: Jev reflex decisions with optional System 2 strategy.Sun, 20 Sep 2026 20:54:32 GMT<p>Multi-drone lab: Jev reflex decisions with optional System 2 strategy.</p><p>An interactive multi-drone autonomy simulation for exploring how fast typed System 1 reflexes can request slower System 2 strategic advice without giving up control. Fleet size is configurable from one to fifteen drones.</p><p><a href="https://github.com/khordoo/jev-reflex-autonomy-lab">khordoo/jev-reflex-autonomy-lab</a></p>Agent Decisionshttps://jevbooks.com/en/p/karanb192__jev-architecthttps://jevbooks.com/en/p/karanb192__jev-architectDesign and evaluate TypeSafe Jev decision loops.Sun, 20 Sep 2026 20:54:31 GMT<p>Design and evaluate TypeSafe Jev decision loops.</p><p>An agent skill that helps you decide what to build with Jev, where it could improve existing software, and what to test before committing. For developers and teams exploring Jev use cases.</p><p><a href="https://github.com/karanb192/jev-architect">karanb192/jev-architect</a></p>Agent Decisionshttps://jevbooks.com/en/p/bodila51__grok-bot-jevhttps://jevbooks.com/en/p/bodila51__grok-bot-jevCheap Jev decision layer for Grok Bot before expensive work.Sun, 20 Sep 2026 20:54:28 GMT<p>Cheap Jev decision layer for Grok Bot before expensive work.</p><p>A small open-source Python reference implementation that connects TypeSafe Jev to Grok Bot as a cheap decision layer, classifying requests before expensive research, browser, retry, or subagent work. For developers integrating Jev with Grok Bot.</p><p><a href="https://github.com/Bodila51/grok-bot-jev">Bodila51/grok-bot-jev</a></p>Agent Decisionsscreen-then-deephttps://jevbooks.com/en/p/burnigtm__jev-mcphttps://jevbooks.com/en/p/burnigtm__jev-mcpLocal stdio MCP server giving Cursor and Codex typed Jev judgments.Sun, 20 Sep 2026 20:54:28 GMT<p>Local stdio MCP server giving Cursor and Codex typed Jev judgments.</p><p>A local stdio MCP server that gives Cursor, Codex, and other MCP clients typed TypeSafe Jev judgments. The host agent still edits files and runs commands; Jev only returns Choice, Score, and Noul answers.</p><p><a href="https://github.com/burnigtm/jev-mcp">burnigtm/jev-mcp</a></p>Agent Decisionshttps://jevbooks.com/en/p/win4r__jev-skill-suggesterhttps://jevbooks.com/en/p/win4r__jev-skill-suggesterBounded installed-skill recommendations with TypeSafe Jev.Sun, 20 Sep 2026 20:54:28 GMT<p>Bounded installed-skill recommendations with TypeSafe Jev.</p><p>A Codex Skill and standalone Python CLI that recommends which installed skill to read first for a task, for agents with many overlapping skills. It does not execute or install skills.</p><p><a href="https://github.com/win4r/jev-skill-suggester">win4r/jev-skill-suggester</a></p>Agent Decisionstiered-thresholdshttps://jevbooks.com/en/p/zeredy879__minojevhttps://jevbooks.com/en/p/zeredy879__minojevTyped calibrated decisions from hidden states, zero output tokens.Sun, 20 Sep 2026 20:54:28 GMT<p>Typed calibrated decisions from hidden states, zero output tokens.</p><p>minojev is an open replica of a decision model that answers runtime-defined Choice, Boolean, and Score questions with calibrated probability distributions in one forward pass, for developers needing fast, parse-free decisions.</p><p><a href="https://github.com/zeredy879/minojev">zeredy879/minojev</a></p>Classification & Routingmeasured-calibrationhttps://jevbooks.com/en/p/keltokhy__jlinkhttps://jevbooks.com/en/p/keltokhy__jlinkRecord linkage with plain-English match rules and Jev probabilities.Sun, 20 Sep 2026 20:54:24 GMT<p>Record linkage with plain-English match rules and Jev probabilities.</p><p>jlink links records for applied economists using a plain-English match rule. It works from Python, CLI, Stata and R on .csv, .dta and .parquet, returning a probability per candidate pair.</p><p><a href="https://github.com/keltokhy/jlink">keltokhy/jlink</a></p>Data Labeling & Curationhttps://jevbooks.com/en/p/keltokhy__jsorthttps://jevbooks.com/en/p/keltokhy__jsortSort lines by a plain-English criterion using Jev pairwise comparisons.Sun, 20 Sep 2026 20:54:24 GMT<p>Sort lines by a plain-English criterion using Jev pairwise comparisons.</p><p>jsort is a Python command-line tool that orders lines, paragraphs, or whole files along a plain-English dimension. It is for anyone who needs to rank or measure text by a qualitative criterion.</p><p><a href="https://github.com/keltokhy/jsort">keltokhy/jsort</a></p>Scoring & Rankingcode-owned-policyhttps://jevbooks.com/en/p/keltokhy__jselecthttps://jevbooks.com/en/p/keltokhy__jselectSource-linked evidence selection within a token budget using Jev Noul relevance.Sun, 20 Sep 2026 20:54:24 GMT<p>Source-linked evidence selection within a token budget using Jev Noul relevance.</p><p>jselect is a Python tool and package that selects relevant, source-linked passages from files, records, or agents, assembling a context within a token budget for AI agents. It is separate from jgrep.</p><p><a href="https://github.com/keltokhy/jselect">keltokhy/jselect</a></p>Scoring & Rankinghttps://jevbooks.com/en/p/lukstei__slop-graderhttps://jevbooks.com/en/p/lukstei__slop-graderRule-based slop grader for text files, powered by Jev.Sun, 20 Sep 2026 20:54:24 GMT<p>Rule-based slop grader for text files, powered by Jev.</p><p>slop-grader is a TypeScript CLI that grades text files against custom or built-in rulesets for AI slop, grammar, and technical doc quality, then helps an AI agent auto-fix violations.</p><p><a href="https://github.com/lukstei/slop-grader">lukstei/slop-grader</a></p>Evaluation & Benchmarkinghttps://jevbooks.com/en/p/shinpr__jev-rerankerhttps://jevbooks.com/en/p/shinpr__jev-rerankerRerank, filter, and compress JSON search results with Jev.Sun, 20 Sep 2026 20:54:20 GMT<p>Rerank, filter, and compress JSON search results with Jev.</p><p>A CLI that takes a JSON array of candidate documents from stdin and outputs a JSON array to stdout, using Jev to rerank, filter, or compress them for LLM context.</p><p><a href="https://github.com/shinpr/jev-reranker">shinpr/jev-reranker</a></p>Scoring & Rankinghttps://jevbooks.com/en/p/uehaj__jev-semgrephttps://jevbooks.com/en/p/uehaj__jev-semgrepGrep by meaning across languages using Jev probabilities.Sun, 20 Sep 2026 20:54:20 GMT<p>Grep by meaning across languages using Jev probabilities.</p><p>semgrep is a command-line tool that finds lines by what they mean, not by regular expressions. It is useful for mixed-language logs, ticket dumps, and teams querying in different languages.</p><p><a href="https://github.com/uehaj/jev-semgrep">uehaj/jev-semgrep</a></p>Scoring & Rankinghttps://jevbooks.com/en/p/minilv__jev-auto-routerhttps://jevbooks.com/en/p/minilv__jev-auto-routerPer-call Codex GPT routing via TypeSafe Jev, with independent task verification.Sun, 20 Sep 2026 20:54:20 GMT<p>Per-call Codex GPT routing via TypeSafe Jev, with independent task verification.</p><p>A prototype for coding agents: a local Responses proxy that lets one Codex session re-select a GPT model and reasoning effort on every meaningful call, then independently verifies whether the task still passed.</p><p><a href="https://github.com/miniLV/Jev-Auto-Router">miniLV/Jev-Auto-Router</a></p>Classification & Routingfailure-mode-knobstate-minimalismhttps://jevbooks.com/en/p/yehuitang0316__jev-nlgrephttps://jevbooks.com/en/p/yehuitang0316__jev-nlgrepNatural-language grep for code, docs, logs, and text.Sun, 20 Sep 2026 20:54:20 GMT<p>Natural-language grep for code, docs, logs, and text.</p><p>nlgrep is a developer tool that searches code, docs, logs, and text by meaning using natural-language queries, finding paraphrases and behavior descriptions even when exact words are forgotten.</p><p><a href="https://github.com/YehuiTang0316/jev-nlgrep">YehuiTang0316/jev-nlgrep</a></p>Scoring & Rankinghttps://jevbooks.com/en/p/virattt__ai-hedge-fundhttps://jevbooks.com/en/p/virattt__ai-hedge-fundEducational AI hedge fund proof of concept with investor agents.Sun, 20 Sep 2026 08:56:00 GMT<p>Educational AI hedge fund proof of concept with investor agents.</p><p>A proof of concept exploring AI-driven trading decisions for educational and research purposes. It runs investor agents over stocks to build and backtest a simulated &quot;fund&quot; — it does not actually place trades.</p><p><a href="https://github.com/virattt/ai-hedge-fund">virattt/ai-hedge-fund</a></p>Finance & Tradinghttps://jevbooks.com/en/p/yangzhou-chaofan__awesome-jev-prompthttps://jevbooks.com/en/p/yangzhou-chaofan__awesome-jev-promptCommunity patterns and top builds for Jev, TypeSafe AI's System One model.Sun, 20 Sep 2026 08:55:54 GMT<p>Community patterns and top builds for Jev, TypeSafe AI's System One model.</p><p>An unofficial, community-maintained collection of prompts, states, and decision patterns for Jev, plus a ranked showcase of community builds. For developers exploring Jev's Choice/Score/Noul primitives and the first 72 hours of ecosystem projects.</p><p><a href="https://github.com/yangzhou-chaofan/awesome-jev-prompt">yangzhou-chaofan/awesome-jev-prompt</a></p>Scoring & Rankingflat-fanouthttps://jevbooks.com/en/p/sivletlabs__jev-evalhttps://jevbooks.com/en/p/sivletlabs__jev-evalBenchmark and one-step RL environment for typed decision models.Sun, 20 Sep 2026 08:55:47 GMT<p>Benchmark and one-step RL environment for typed decision models.</p><p>A benchmark and Gymnasium-style one-step bandit for typed decision models like Jev/System One. It evaluates models that map unstructured state and a closed question to a probability distribution over declared candidates, with no free-text generation.</p><p><a href="https://github.com/SivletLabs/jev-eval">SivletLabs/jev-eval</a></p>Evaluation & Benchmarkinghttps://jevbooks.com/en/p/systemgroupnet__openwebui-ask-jevhttps://jevbooks.com/en/p/systemgroupnet__openwebui-ask-jevOpen WebUI Tool that sends typed questions to Jev.Sun, 20 Sep 2026 08:55:36 GMT<p>Open WebUI Tool that sends typed questions to Jev.</p><p>An Open WebUI Tool that lets a chat model send typed questions to TypeSafe's Jev and receive calibrated probabilities instead of guesses. Aimed at users wanting defensible scores, fast structured decisions, batch labelling, and rubric grading.</p><p><a href="https://github.com/systemgroupnet/openwebui-ask-jev">systemgroupnet/openwebui-ask-jev</a></p>Classification & Routingflat-fanouthttps://jevbooks.com/en/p/angelgalvisc__snake-arena-jev-vs-llmshttps://jevbooks.com/en/p/angelgalvisc__snake-arena-jev-vs-llmsJev vs six LLMs: decisions per minute and cost in Snake.Sun, 20 Sep 2026 08:55:32 GMT<p>Jev vs six LLMs: decisions per minute and cost in Snake.</p><p>A benchmark racing Jev, TypeSafe's System One decision model, against six general-purpose LLMs on Snake. Each player gets sixty seconds and its own board; the repo measures decisions per minute and cost per point.</p><p><a href="https://github.com/angelgalvisc/snake-arena-jev-vs-llms">angelgalvisc/snake-arena-jev-vs-llms</a></p>Evaluation & Benchmarkinghttps://jevbooks.com/en/p/beizi6__jev-introductionhttps://jevbooks.com/en/p/beizi6__jev-introductionChinese introduction to Jev: System One, State, Choice, Score, Noul.Sun, 20 Sep 2026 08:55:22 GMT<p>Chinese introduction to Jev: System One, State, Choice, Score, Noul.</p><p>A Chinese-language article based on TypeSafe's official docs, explaining Jev's System One model, State, Choice, Score, Noul, structured questions, probabilities, confidence, and practical limits. Available as Markdown, HTML, and PDF for Chinese-speaking readers.</p><p><a href="https://github.com/BeiZi6/jev-introduction">BeiZi6/jev-introduction</a></p>Calibration & Researchhttps://jevbooks.com/en/p/sleepy0x13__jevia-harnesshttps://jevbooks.com/en/p/sleepy0x13__jevia-harnessA local harness that puts Jev in front of the LLM.Sun, 20 Sep 2026 08:55:11 GMT<p>A local harness that puts Jev in front of the LLM.</p><p>A local agent harness for people running tasks that are mostly typed judgements. Jev handles every judgement expressible as a type; an ordinary LLM is paid for only when prose is actually needed.</p><p><a href="https://github.com/sleepy0x13/jevia-harness">sleepy0x13/jevia-harness</a></p>Classification & Routingflat-fanoutscreen-then-deephttps://jevbooks.com/en/p/parth-kp__jev-mail-classifierhttps://jevbooks.com/en/p/parth-kp__jev-mail-classifierClassify your inbox with Jev — tag, move, flag, notify.Sun, 20 Sep 2026 08:55:06 GMT<p>Classify your inbox with Jev — tag, move, flag, notify.</p><p>A Python terminal tool that connects to your IMAP mailbox, classifies unprocessed email against plain-language categories, and runs config-driven actions like tag, move, flag, or webhook. Aimed at anyone who wants inbox automation without prompt engineering.</p><p><a href="https://github.com/parth-kp/jev-mail-classifier">parth-kp/jev-mail-classifier</a></p>Classification & Routingflat-fanouttiered-thresholdshttps://jevbooks.com/en/p/supremedreamz__jev-fastloophttps://jevbooks.com/en/p/supremedreamz__jev-fastloopCheap fail-open semantic edge layer for Jev decisions.Sun, 20 Sep 2026 08:54:59 GMT<p>Cheap fail-open semantic edge layer for Jev decisions.</p><p>A stdlib-only Python decision service for developers using Jev (TypeSafe System One). It wraps typed questions with a confidence policy, per-lane decision logging, a blind eval harness, and a skill router over large skill rosters.</p><p><a href="https://github.com/SupremeDreamZ/jev-fastloop">SupremeDreamZ/jev-fastloop</a></p>Classification & Routingflat-fanouthttps://jevbooks.com/en/p/ashadeepa__typesafe-showcasehttps://jevbooks.com/en/p/ashadeepa__typesafe-showcaseNext.js UI showcasing TypeSafe's System One (Jev) model use cases.Sun, 20 Sep 2026 08:54:59 GMT<p>Next.js UI showcasing TypeSafe's System One (Jev) model use cases.</p><p>A Next.js UI for showing and telling TypeSafe System One (Jev) use cases, deployable to Vercel. Bring your own key: visitors paste their own TYPESAFE_API_KEY, stored only in localStorage.</p><p><a href="https://github.com/Ashadeepa/typesafe-showcase">Ashadeepa/typesafe-showcase</a></p>Evaluation & Benchmarkinghttps://jevbooks.com/en/p/ash20pk__beat-the-reviewerhttps://jevbooks.com/en/p/ash20pk__beat-the-reviewerTen approvals to win, judged by a probability, not a sentence.Sun, 20 Sep 2026 08:54:53 GMT<p>Ten approvals to win, judged by a probability, not a sentence.</p><p>A game where you write pull request bodies to convince a reviewer to approve ten risky production requests. For developers who want to practice precise, scoped arguments under a probabilistic gate.</p><p><a href="https://github.com/Ash20pk/beat-the-reviewer">Ash20pk/beat-the-reviewer</a></p>Verification & Guardrailshttps://jevbooks.com/en/p/mawfyy__jev-router.nvimhttps://jevbooks.com/en/p/mawfyy__jev-router.nvimNeovim intent router classifying AI commands with Jev.Sun, 20 Sep 2026 08:54:46 GMT<p>Neovim intent router classifying AI commands with Jev.</p><p>A Neovim plugin that classifies AI commands using Jev via the OpenRouter Decisions API and routes execution to matching handlers for file reading, code editing, command running, or general questions.</p><p><a href="https://github.com/Mawfyy/jev-router.nvim">Mawfyy/jev-router.nvim</a></p>Classification & Routingcandidate-enumerationhttps://jevbooks.com/en/p/2951461586__jev-register-toolhttps://jevbooks.com/en/p/2951461586__jev-register-toolAutomates TypeSafe (Jev) account registration and API key creation via pure HTTP.Sun, 20 Sep 2026 08:54:46 GMT<p>Automates TypeSafe (Jev) account registration and API key creation via pure HTTP.</p><p>A Python tool for automating the full TypeSafe (Jev / System One) account lifecycle: apply, confirm email, wait for approval, register, create API key, and store in a ledger. It uses pure HTTP with no browser, and can batch-produce accounts.</p><p><a href="https://github.com/2951461586/Jev-Register-Tool">2951461586/Jev-Register-Tool</a></p>Infra / SDKs / Integrationshttps://jevbooks.com/en/p/abdelstark__awesome-typesafehttps://jevbooks.com/en/p/abdelstark__awesome-typesafeCurated list of TypeSafe, System One, and Jev resources.Sun, 20 Sep 2026 08:54:32 GMT<p>Curated list of TypeSafe, System One, and Jev resources.</p><p>A curated list of official resources and community projects for TypeSafe, System One models, and Jev, for developers building with typed probabilistic decisions.</p><p><a href="https://github.com/AbdelStark/awesome-typesafe">AbdelStark/awesome-typesafe</a></p>Calibration & Researchhttps://jevbooks.com/en/p/zhangxaochen__dsh-jevhttps://jevbooks.com/en/p/zhangxaochen__dsh-jevJev decision primitives as a Cordis plugin suite for DeepSeek Harness.Sun, 20 Sep 2026 08:54:20 GMT<p>Jev decision primitives as a Cordis plugin suite for DeepSeek Harness.</p><p>A Cordis plugin suite pairing Jev (TypeSafe System One decision models) with DeepSeek Harness (dsh), adding semantic judgement dsh lacks: dynamic tool pruning, semantic dead-loop blocking, and a guard for high-risk execution.</p><p><a href="https://github.com/zhangxaochen/dsh-jev">zhangxaochen/dsh-jev</a></p>Agent Decisionsflat-fanouthttps://jevbooks.com/en/p/opsiaadi__opnroute-jevhttps://jevbooks.com/en/p/opsiaadi__opnroute-jevMCP server exposing JEV classification tools via OpenRouter.Sun, 20 Sep 2026 08:54:09 GMT<p>MCP server exposing JEV classification tools via OpenRouter.</p><p>A Python MCP server that classifies text using JEV (TypeSafe System One) through OpenRouter's Streamable HTTP API, for Hermes and other MCP clients such as Claude Desktop and Cursor.</p><p><a href="https://github.com/opsiaadi/opnroute-jev">opsiaadi/opnroute-jev</a></p>Classification & Routinghttps://jevbooks.com/en/p/wh000wh000__awesome-jev-livehttps://jevbooks.com/en/p/wh000wh000__awesome-jev-liveEvidence-graded index of the Jev / TypeSafe System One ecosystem.Sun, 20 Sep 2026 08:54:03 GMT<p>Evidence-graded index of the Jev / TypeSafe System One ecosystem.</p><p>An auto-rebuilding index of projects built on Jev, TypeSafe AI's System One model. Entries are collected, filtered and re-checked by a pipeline every two hours across 20 languages, graded by evidence level for developers exploring the ecosystem.</p><p><a href="https://github.com/wh000wh000/awesome-jev-live">wh000wh000/awesome-jev-live</a></p>Data Labeling & Curationhttps://jevbooks.com/en/p/sawzhang__jev-demohttps://jevbooks.com/en/p/sawzhang__jev-demoHands-on Jev study: concepts, 5 runnable demos, reproducible benchmarks.Sun, 20 Sep 2026 08:53:52 GMT<p>Hands-on Jev study: concepts, 5 runnable demos, reproducible benchmarks.</p><p>A learning and benchmarking record for Jev (TypeSafe System One), the judgment model that returns typed probabilistic answers instead of free text. Aimed at developers evaluating Jev for triage, guardrails, reranking, or function calling.</p><p><a href="https://github.com/sawzhang/jev-demo">sawzhang/jev-demo</a></p>Calibration & Researchflat-fanoutspeculative-headshttps://jevbooks.com/en/p/xjfyt__jev-cpu-demohttps://jevbooks.com/en/p/xjfyt__jev-cpu-demoUnofficial CPU-first System One–compatible decision server with Playwright browser agent demo.Sun, 20 Sep 2026 08:53:45 GMT<p>Unofficial CPU-first System One–compatible decision server with Playwright browser agent demo.</p><p>A local FastAPI decision server plus a Playwright Chromium agent and operator console UI with 10 mainland-China-accessible site templates. For developers exploring browser automation without GPU or TypeSafe calls.</p><p><a href="https://github.com/xjfyt/jev-cpu-demo">xjfyt/jev-cpu-demo</a></p>Agent Decisionshttps://jevbooks.com/en/p/cobusgreyling__jevhttps://jevbooks.com/en/p/cobusgreyling__jevUnofficial TypeSafe Jev showcase: System One decisions, not chat.Sun, 20 Sep 2026 08:53:45 GMT<p>Unofficial TypeSafe Jev showcase: System One decisions, not chat.</p><p>An unofficial operator-level showcase for TypeSafe Jev, aimed at developers building classify, route, score, or verify steps. It provides a Python operator lab, a smart-home fan-out demo, a TypeScript harness CLI, and agent skills.</p><p><a href="https://github.com/cobusgreyling/Jev">cobusgreyling/Jev</a></p>Agent Decisionsflat-fanouthttps://jevbooks.com/en/p/talya1412__jev-harnesshttps://jevbooks.com/en/p/talya1412__jev-harnessTypeSafe Jev integrations for agent harnesses, CI, CLIs, and editors.Sun, 20 Sep 2026 08:53:33 GMT<p>TypeSafe Jev integrations for agent harnesses, CI, CLIs, and editors.</p><p>Fourteen TypeScript packages that integrate TypeSafe Jev, a decision model returning calibrated probabilities, into agent harnesses, GitHub Actions, CLIs, and editor extensions for routing, gating, and verification.</p><p><a href="https://github.com/Talya1412/jev-harness">Talya1412/jev-harness</a></p>Classification & Routingflat-fanouthttps://jevbooks.com/en/p/wustep__jev-playgroundhttps://jevbooks.com/en/p/wustep__jev-playgroundJev picks enum labels; code renders sheet, audio and MIDI.Sun, 20 Sep 2026 08:53:30 GMT<p>Jev picks enum labels; code renders sheet, audio and MIDI.</p><p>A TypeScript playground testing whether TypeSafe's System One model Jev can steer music composition across eight styles, plus an absurd trolley-problem demo, for developers exploring typed model decisions.</p><p><a href="https://github.com/wustep/jev-playground">wustep/jev-playground</a></p>Game & Simulationhttps://jevbooks.com/en/p/licensedsaucer9-web__jev-opportunitieshttps://jevbooks.com/en/p/licensedsaucer9-web__jev-opportunitiesJev opportunity map for Grok Bot and Grok Build builders.Sun, 20 Sep 2026 08:53:22 GMT<p>Jev opportunity map for Grok Bot and Grok Build builders.</p><p>A research repo for builders of Grok Bot and Grok Build who want to offload System One decisions to TypeSafe Jev or free/open models, and to build Jev-powered products with those agents.</p><p><a href="https://github.com/licensedsaucer9-web/jev-opportunities">licensedsaucer9-web/jev-opportunities</a></p>Agent Decisionshttps://jevbooks.com/en/p/llt22__jev-labhttps://jevbooks.com/en/p/llt22__jev-labIndependent lab for Jev: benchmarks, use cases, and ecosystem audit.Sun, 20 Sep 2026 08:53:18 GMT<p>Independent lab for Jev: benchmarks, use cases, and ecosystem audit.</p><p>A community-run research lab for TypeSafe's Jev (System One model). It curates official resources, SDKs, and real-world use cases, and provides reproducible benchmarks of Noul, Choice, and Score primitives, confidence gating, fan-out latency, and agent control.</p><p><a href="https://github.com/llt22/jev-lab">llt22/jev-lab</a></p>Evaluation & Benchmarkinghttps://jevbooks.com/en/p/kaustav1996__reflexhttps://jevbooks.com/en/p/kaustav1996__reflexCoding agent with calibrated System One reflexes on top of PiSun, 20 Sep 2026 08:53:15 GMT<p>Coding agent with calibrated System One reflexes on top of Pi</p><p>A coding agent and personal assistant built on the Pi coding agent. It works with any LLM via OpenRouter or a direct provider, takes voice input, and includes macOS computer use, a browser agent, MCP connectors, scheduled agents and deploys.</p><p><a href="https://github.com/kaustav1996/reflex">kaustav1996/reflex</a></p>Verification & Guardrailscross-checktiered-thresholdshttps://jevbooks.com/en/p/omnijev__awesome-jev-galleryhttps://jevbooks.com/en/p/omnijev__awesome-jev-galleryPapers, open models and evaluations behind System One models and Jev.Sun, 20 Sep 2026 08:53:11 GMT<p>Papers, open models and evaluations behind System One models and Jev.</p><p>A curated awesome list of papers, open reproductions and independent evaluations behind System One models and Jev, for researchers and engineers tracking the typed-decision ecosystem.</p><p><a href="https://github.com/OmniJev/awesome-jev-gallery">OmniJev/awesome-jev-gallery</a></p>Evaluation & Benchmarkinghttps://jevbooks.com/en/p/pantos12__mailverdicthttps://jevbooks.com/en/p/pantos12__mailverdictForward an email, get a calibrated phishing verdict.Sun, 20 Sep 2026 08:52:56 GMT<p>Forward an email, get a calibrated phishing verdict.</p><p>MailVerdict parses an email deterministically, asks a calibrated classifier one question set, and returns a phishing probability plus indicators. It ships as an MCP server for Copilot Studio and a REST endpoint for Power Automate.</p><p><a href="https://github.com/pantos12/mailverdict">pantos12/mailverdict</a></p>Classification & Routinghttps://jevbooks.com/en/p/reachjalil__system-one-benchhttps://jevbooks.com/en/p/reachjalil__system-one-benchBenchmark catalog for Jev evidence, findings, and reproducible decision benchmarks.Sun, 20 Sep 2026 08:52:56 GMT<p>Benchmark catalog for Jev evidence, findings, and reproducible decision benchmarks.</p><p>A public catalog of experiments measuring what Jev can help an agent do, for individual agent users and builders. It records each experiment's measurement, baseline, and conclusion limits.</p><p><a href="https://github.com/reachjalil/system-one-bench">reachjalil/system-one-bench</a></p>Evaluation & Benchmarkinghttps://jevbooks.com/en/p/taku-me__chakuhohttps://jevbooks.com/en/p/taku-me__chakuhoJev-compatible System One decision endpoint over a local LLM.Sun, 20 Sep 2026 08:52:45 GMT<p>Jev-compatible System One decision endpoint over a local LLM.</p><p>chakuho is a Python HTTP service that answers typed decision questions (choice, noul/boolean, score) about a given state, returning probability distributions instead of generated text. It targets developers who want Jev-style System One decisions from an already-running local instruct model.</p><p><a href="https://github.com/taku-me/chakuho">taku-me/chakuho</a></p>Agent Decisionsflat-fanouthttps://jevbooks.com/en/p/uspraveen__jevifyhttps://jevbooks.com/en/p/uspraveen__jevifyTurn any open LLM into a calibrated System One decision model.Sun, 20 Sep 2026 08:52:45 GMT<p>Turn any open LLM into a calibrated System One decision model.</p><p>Jevify is an open-source benchmark, engine, and training recipe that gives any Hugging Face checkpoint the same typed-question interface as TypeSafe's Jev, for developers who need calibrated probability distributions instead of generated text.</p><p><a href="https://github.com/uspraveen/Jevify">uspraveen/Jevify</a></p>Calibration & Researchmeasured-calibrationhttps://jevbooks.com/en/p/itisshikhar__gg-friggin-ezhttps://jevbooks.com/en/p/itisshikhar__gg-friggin-ezDrop-in profanity and toxicity screener for Node.js, powered by Jev.Sun, 20 Sep 2026 08:52:40 GMT<p>Drop-in profanity and toxicity screener for Node.js, powered by Jev.</p><p>A Node.js profanity and toxicity screener for real-time chat and game moderation. It detects leetspeak, character spacing, romanized and code-mixed profanity across English and Indic languages including Kannada, Telugu, Tamil, Hindi, and Bengali.</p><p><a href="https://github.com/ItisShikhar/gg-friggin-ez">ItisShikhar/gg-friggin-ez</a></p>Content Moderationhttps://jevbooks.com/en/p/ymik-kimy__lifeinboxhttps://jevbooks.com/en/p/ymik-kimy__lifeinboxLightweight daily-planning assistant using Jev to evaluate tasks.Sun, 20 Sep 2026 08:52:34 GMT<p>Lightweight daily-planning assistant using Jev to evaluate tasks.</p><p>LifeInbox is a lightweight daily-planning assistant for individuals. It evaluates each manually entered task by importance, urgency, energy requirement, duration, and postponability, then organizes tasks into today's priorities, quick actions, optional tasks, and items requiring confirmation.</p><p><a href="https://github.com/ymik-kimy/LifeInbox">ymik-kimy/LifeInbox</a></p>Scoring & Rankinghttps://jevbooks.com/en/p/wallerchen__jev-measuredhttps://jevbooks.com/en/p/wallerchen__jev-measuredMeasured cost, latency, and raw output from the live Jev API across eight use cases.Sun, 20 Sep 2026 08:52:28 GMT<p>Measured cost, latency, and raw output from the live Jev API across eight use cases.</p><p>A reproducible benchmark measuring cost, latency, and raw output from the live Jev API (TypeSafe AI's System One decision model) across eight realistic use cases, for developers evaluating Jev.</p><p><a href="https://github.com/WallerChen/jev-measured">WallerChen/jev-measured</a></p>Evaluation & Benchmarkingflat-fanouthttps://jevbooks.com/en/p/lychoon__paper-radar-jevhttps://jevbooks.com/en/p/lychoon__paper-radar-jevDaily arXiv paper radar scored by TypeSafe AI relevance.Sun, 20 Sep 2026 08:52:22 GMT<p>Daily arXiv paper radar scored by TypeSafe AI relevance.</p><p>A configurable Python CLI for researchers: it fetches recent arXiv papers, evaluates how well each matches a JSON research profile, and writes daily Markdown and JSON relevance rankings.</p><p><a href="https://github.com/LYchoon/paper-radar-jev">LYchoon/paper-radar-jev</a></p>Scoring & Rankinghttps://jevbooks.com/en/p/juanpiriv__usher-pointhttps://jevbooks.com/en/p/juanpiriv__usher-pointCLI that decides where a task runs and dispatches it.Sun, 20 Sep 2026 08:52:18 GMT<p>CLI that decides where a task runs and dispatches it.</p><p>usher-point is an on-demand CLI for developers that decides where a task should run — inline in Claude Code, via Codex CLI in a sandbox, or in an isolated Orca worktree — and optionally which skills and model to use, then dispatches it.</p><p><a href="https://github.com/juanpiRiv/usher-point">juanpiRiv/usher-point</a></p>Classification & Routingfailure-mode-knobhttps://jevbooks.com/en/p/thiagoadril__typesafe-docshttps://jevbooks.com/en/p/thiagoadril__typesafe-docsJev and System One documentation extracted into markdown for LLM training.Sun, 20 Sep 2026 08:52:05 GMT<p>Jev and System One documentation extracted into markdown for LLM training.</p><p>A repository of markdown documents extracted from the typesafe.ai documentation site. It serves as a reference index of Jev and System One material intended to be consumed by LLMs or used for LLM training.</p><p><a href="https://github.com/thiagoadril/typesafe-docs">thiagoadril/typesafe-docs</a></p>Otherhttps://jevbooks.com/en/p/rhc98__awesome-jevhttps://jevbooks.com/en/p/rhc98__awesome-jevCurated list of projects built on Jev, judged by Jev itself.Sun, 20 Sep 2026 08:51:59 GMT<p>Curated list of projects built on Jev, judged by Jev itself.</p><p>A community-curated index of open-source projects built on Jev, TypeSafe AI's System One model. Each repository was collected from GitHub and judged by Jev in a single call.</p><p><a href="https://github.com/rhc98/awesome-jev">rhc98/awesome-jev</a></p>Data Labeling & Curationhttps://jevbooks.com/en/p/sontakey__awesome-jevhttps://jevbooks.com/en/p/sontakey__awesome-jevUnofficial list of projects using TypeSafe AI Jev.Sun, 20 Sep 2026 08:51:56 GMT<p>Unofficial list of projects using TypeSafe AI Jev.</p><p>An unofficial, curated catalogue of open-source projects that use Jev, TypeSafe AI's System One model, for developers looking for real-world examples of Choice, Score, and Noul usage.</p><p><a href="https://github.com/sontakey/awesome-jev">sontakey/awesome-jev</a></p>Classification & Routingcandidate-enumerationhttps://jevbooks.com/en/p/minorun365__jev-cloud-quizhttps://jevbooks.com/en/p/minorun365__jev-cloud-quizJev classifies cloud service names with probabilities.Sun, 20 Sep 2026 08:51:49 GMT<p>Jev classifies cloud service names with probabilities.</p><p>A demo where Jev, TypeSafe AI's System One model, determines which of three major clouds (AWS, Azure, Google Cloud) a given service name belongs to, showing probabilities. For developers exploring probabilistic classification.</p><p><a href="https://github.com/minorun365/jev-cloud-quiz">minorun365/jev-cloud-quiz</a></p>Classification & Routinghttps://jevbooks.com/en/p/tincke10__jevesthttps://jevbooks.com/en/p/tincke10__jevestAutomated PR review pipeline using Jev as a millisecond decision layer.Sun, 20 Sep 2026 08:51:44 GMT<p>Automated PR review pipeline using Jev as a millisecond decision layer.</p><p>Jevest is an automated PR review pipeline for developers. It uses Jev (TypeSafe AI) as a millisecond decision layer over an LLM reviewer to decide what to review, how much, and what to publish. Status: phase 0/1 skeleton.</p><p><a href="https://github.com/tincke10/Jevest">tincke10/Jevest</a></p>Classification & Routinghttps://jevbooks.com/en/p/kraayenjon__awesome-jevhttps://jevbooks.com/en/p/kraayenjon__awesome-jevCurated list of Jev projects, SDKs, tools, and learning resources.Sun, 20 Sep 2026 08:51:36 GMT<p>Curated list of Jev projects, SDKs, tools, and learning resources.</p><p>An unofficial curated list of use cases, projects, SDKs, tools, and learning resources for Jev, TypeSafe AI's System One model for typed decisions inside software. It helps developers explore the early ecosystem.</p><p><a href="https://github.com/kraayenjon/awesome-jev">kraayenjon/awesome-jev</a></p>Otherflat-fanouthttps://jevbooks.com/en/p/twilwa__pi-typesafehttps://jevbooks.com/en/p/twilwa__pi-typesafeTypeSafe Jev sidecar adds pre-flight and diff-quality hazard checks to Pi.Sun, 20 Sep 2026 08:51:35 GMT<p>TypeSafe Jev sidecar adds pre-flight and diff-quality hazard checks to Pi.</p><p>A Pi coding-agent extension that runs four pre-flight hazard checks before bash, write and edit, and four diff-quality checks after write and edit, generating fixed advisory feedback for the agent.</p><p><a href="https://github.com/twilwa/pi-typesafe">twilwa/pi-typesafe</a></p>Verification & Guardrailshttps://jevbooks.com/en/p/sagochiko__jev-lateral-thinking-quizhttps://jevbooks.com/en/p/sagochiko__jev-lateral-thinking-quizJev as quizmaster for lateral thinking puzzlesSun, 20 Sep 2026 08:51:26 GMT<p>Jev as quizmaster for lateral thinking puzzles</p><p>A web app where players ask free-form Japanese questions to solve lateral thinking puzzles, and Jev answers yes/no/irrelevant and judges final answers.</p><p><a href="https://github.com/sagochiko/jev-lateral-thinking-quiz">sagochiko/jev-lateral-thinking-quiz</a></p>Game & Simulationhttps://jevbooks.com/en/p/jumboly__jev-map-sandboxhttps://jevbooks.com/en/p/jumboly__jev-map-sandboxTesting Jev's typed judgments on map layer importanceSun, 20 Sep 2026 08:51:14 GMT<p>Testing Jev's typed judgments on map layer importance</p><p>A browser-only sandbox experimenting with Jev on maps. Experiment 01 classifies the importance of 24 groups from the 123-layer 地理院地図Vector style catalog based on spoken instructions, then reflects the result on a MapLibre GL JS map.</p><p><a href="https://github.com/jumboly/jev-map-sandbox">jumboly/jev-map-sandbox</a></p>Scoring & Rankingflat-fanouttiered-thresholdshttps://jevbooks.com/en/p/julardepick__jev-examinerhttps://jevbooks.com/en/p/julardepick__jev-examinerLayered AI content moderation workflow powered by TypeSafe/Jev.Sun, 20 Sep 2026 08:51:09 GMT<p>Layered AI content moderation workflow powered by TypeSafe/Jev.</p><p>A general-purpose AI content moderation workflow for developers and operators. It decomposes a broad violation judgment into atomic questions, combines weighted scores, and routes items to pass, human review, or block.</p><p><a href="https://github.com/JularDepick/Jev-Examiner">JularDepick/Jev-Examiner</a></p>Content Moderationflat-fanoutcode-owned-policyhttps://jevbooks.com/en/p/ordepas__system1-fraud-interceptor-demohttps://jevbooks.com/en/p/ordepas__system1-fraud-interceptor-demoSimulated fraud interceptor comparing Jev and Gemini on synthetic transactions.Sun, 20 Sep 2026 08:51:02 GMT<p>Simulated fraud interceptor comparing Jev and Gemini on synthetic transactions.</p><p>A personal experimentation demo that runs a simulated fraud interceptor over synthetic transactions, comparing a fast typed-decision model (Jev) against an LLM (Gemini) in parallel on latency, token usage and decisions. Not a benchmark.</p><p><a href="https://github.com/ordepas/system1-fraud-interceptor-demo">ordepas/system1-fraud-interceptor-demo</a></p>Finance & Tradinghttps://jevbooks.com/en/p/furucrm-inc__flash-agent-stackhttps://jevbooks.com/en/p/furucrm-inc__flash-agent-stackLocal open-source drop-in for TypeSafe AI Jev with WebMCP bridge.Sun, 20 Sep 2026 08:50:59 GMT<p>Local open-source drop-in for TypeSafe AI Jev with WebMCP bridge.</p><p>A local, free, Jev-compatible System 1 AI stack for developers and Salesforce teams: a backend API, Salesforce LWC demo, and a Chrome extension that injects document.modelContext into any web page.</p><p><a href="https://github.com/furuCRM-Inc/flash-agent-stack">furuCRM-Inc/flash-agent-stack</a></p>Infra / SDKs / Integrationsflat-fanouthttps://jevbooks.com/en/p/tpellet__grevihttps://jevbooks.com/en/p/tpellet__grevigrep for meaning: a Unix CLI that searches by description.Sun, 20 Sep 2026 08:50:49 GMT<p>grep for meaning: a Unix CLI that searches by description.</p><p>grevi is a Rust command-line tool that searches by meaning: it finds the error in a long log, a commit, a command, or the right folder for a file, from a plain-English description.</p><p><a href="https://github.com/tpellet/grevi">tpellet/grevi</a></p>Scoring & Rankinghttps://jevbooks.com/en/p/momomo-agent__judge-vm-webhttps://jevbooks.com/en/p/momomo-agent__judge-vm-webInteractive playground for JASM v0.2, the assembly language of JLMP.Sun, 20 Sep 2026 08:50:35 GMT<p>Interactive playground for JASM v0.2, the assembly language of JLMP.</p><p>A web playground for JASM v0.2, the assembly language of JLMP. It parses and executes JASM programs, dispatching each of the four primitives (J/L/M/P) to its own backend, with mock modes for local experimentation.</p><p><a href="https://github.com/momomo-agent/judge-vm-web">momomo-agent/judge-vm-web</a></p>Evaluation & Benchmarkinghttps://jevbooks.com/en/p/tickerdev__jevfanity-apihttps://jevbooks.com/en/p/tickerdev__jevfanity-apiProfanity and moderation API using Jev for multi-category scoring.Sun, 20 Sep 2026 08:50:27 GMT<p>Profanity and moderation API using Jev for multi-category scoring.</p><p>Jevfanity is a TypeScript API deployed on Cloudflare Workers that detects profanity and other moderation categories in text. It is for developers who need a hosted or self-hosted moderation endpoint.</p><p><a href="https://github.com/TickerDev/jevfanity-api">TickerDev/jevfanity-api</a></p>Content Moderationflat-fanouthttps://jevbooks.com/en/p/simonsez9510__jev-cite-checkhttps://jevbooks.com/en/p/simonsez9510__jev-cite-checkJev gate checks whether AI answers are grounded in cited ordinance articles.Sun, 20 Sep 2026 08:50:21 GMT<p>Jev gate checks whether AI answers are grounded in cited ordinance articles.</p><p>A reproducibility package for an experiment that asks Jev (TypeSafe AI System One) to judge whether an AI-generated sentence about a local ordinance is actually grounded in the article it cites. For researchers testing a separate scoring gate on generated answers.</p><p><a href="https://github.com/simonsez9510/jev-cite-check">simonsez9510/jev-cite-check</a></p>Verification & Guardrailscross-checkhttps://jevbooks.com/en/p/agarwalpranav0711__jev-ai-testinghttps://jevbooks.com/en/p/agarwalpranav0711__jev-ai-testingTwo Jev-powered web demos: tone analyzer and maze runner.Sun, 20 Sep 2026 08:50:14 GMT<p>Two Jev-powered web demos: tone analyzer and maze runner.</p><p>A suite of two browser demos for developers and demo viewers: Mirror, a real-time message tone analyzer scoring 8 tone dimensions, and Jev Maze Runner, a game where Jev picks paths through a random 15x15 maze.</p><p><a href="https://github.com/agarwalpranav0711/jev-ai-testing">agarwalpranav0711/jev-ai-testing</a></p>Game & Simulationhttps://jevbooks.com/en/p/allenporter__home-assistant-typesafehttps://jevbooks.com/en/p/allenporter__home-assistant-typesafeHome Assistant conversation integration using Jev for fast intent routing.Sun, 20 Sep 2026 08:50:08 GMT<p>Home Assistant conversation integration using Jev for fast intent routing.</p><p>A Home Assistant custom conversation integration powered by TypeSafe AI System One (Jev model family) for fast, structured intent routing and device control, aimed at Home Assistant users.</p><p><a href="https://github.com/allenporter/home-assistant-typesafe">allenporter/home-assistant-typesafe</a></p>Classification & Routinghttps://jevbooks.com/en/p/sanmai__typesafe-ai-phphttps://jevbooks.com/en/p/sanmai__typesafe-ai-phpPHP SDK for TypeSafe AI: ask typed questions, get typed answers.Sun, 20 Sep 2026 08:50:03 GMT<p>PHP SDK for TypeSafe AI: ask typed questions, get typed answers.</p><p>A PHP client for the TypeSafe AI evaluation API. Provide a state and typed questions, get one typed answer per question. For PHP 8.2+ developers integrating Jev into their applications.</p><p><a href="https://github.com/sanmai/typesafe-ai-php">sanmai/typesafe-ai-php</a></p>Evaluation & Benchmarkingflat-fanouthttps://jevbooks.com/en/p/lirantal__discoprinthttps://jevbooks.com/en/p/lirantal__discoprintClassify an artist's discography by theme, mood, and lyrical complexity with Jev.Sun, 20 Sep 2026 08:49:55 GMT<p>Classify an artist's discography by theme, mood, and lyrical complexity with Jev.</p><p>discoprint is a TypeScript CLI tool that classifies an artist's discography by theme, mood, and lyrical complexity using Jev, then displays the results as a colorful terminal dashboard.</p><p><a href="https://github.com/lirantal/discoprint">lirantal/discoprint</a></p>Data Labeling & Curationflat-fanouthttps://jevbooks.com/en/p/bald0wang__jev-docs-zhhttps://jevbooks.com/en/p/bald0wang__jev-docs-zhUnofficial Chinese translation of the official Jev (TypeSafe AI) docs.Sun, 20 Sep 2026 08:49:51 GMT<p>Unofficial Chinese translation of the official Jev (TypeSafe AI) docs.</p><p>A community translation project that renders the official Jev documentation (TypeSafe AI's System One model) into Simplified Chinese, covering all 109 pages, and ships a static site generator to build and host the translated docs.</p><p><a href="https://github.com/Bald0Wang/jev-docs-zh">Bald0Wang/jev-docs-zh</a></p>Otherhttps://jevbooks.com/en/p/arohtea__jev-readouthttps://jevbooks.com/en/p/arohtea__jev-readoutStructured, probability-bearing judgments from Jev, with raw HTTP transparency.Sun, 20 Sep 2026 08:49:50 GMT<p>Structured, probability-bearing judgments from Jev, with raw HTTP transparency.</p><p>A local web tool for developers evaluating TypeSafe AI's Jev model: paste a state, define typed questions, and see each answer with its full probability distribution plus the raw request/response bytes.</p><p><a href="https://github.com/Arohtea/jev-readout">Arohtea/jev-readout</a></p>Evaluation & Benchmarkingflat-fanoutno-judgment-bandhttps://jevbooks.com/en/p/wjdjdakf17__jev-studyhttps://jevbooks.com/en/p/wjdjdakf17__jev-studyKorean study notes and TypeScript mockups for Jev's typed decisionsSun, 20 Sep 2026 08:49:35 GMT<p>Korean study notes and TypeScript mockups for Jev's typed decisions</p><p>A personal Korean-language study repo on Jev, TypeSafe AI's &quot;System One Model&quot;. It turns Jev's published concepts into TypeScript types, a deterministic mock client, a routing workflow, and calibration math, aimed at readers who want to understand the ideas before the API is public.</p><p><a href="https://github.com/wjdjdakf17/jev-study">wjdjdakf17/jev-study</a></p>Classification & Routingmeasured-calibrationhttps://jevbooks.com/en/p/skcache__jevtrafficsimhttps://jevbooks.com/en/p/skcache__jevtrafficsimTop-down traffic simulator comparing Fixed, Adaptive, and Jev signal controllersSun, 20 Sep 2026 08:49:28 GMT<p>Top-down traffic simulator comparing Fixed, Adaptive, and Jev signal controllers</p><p>An interactive top-down browser traffic simulator that compares three traffic-signal controllers — Fixed, Adaptive, and Jev — under identical seeded workloads, with a headless benchmark harness for reproducible runs.</p><p><a href="https://github.com/skcache/jevtrafficsim">skcache/jevtrafficsim</a></p>Game & Simulationhttps://jevbooks.com/en/p/hari31416__typesafe-demo-finance-tracking-apphttps://jevbooks.com/en/p/hari31416__typesafe-demo-finance-tracking-appPersonal finance tracker using Jev's typed decisions for categorization and anomaly detection.Sun, 20 Sep 2026 08:49:20 GMT<p>Personal finance tracker using Jev's typed decisions for categorization and anomaly detection.</p><p>A demo personal finance tracker with a Python/FastAPI backend and React/TypeScript frontend. It categorizes transactions, detects subscriptions and duplicates, and flags anomalies using Jev's calibrated probabilities, then applies deterministic budget rules.</p><p><a href="https://github.com/Hari31416/typesafe-demo-finance-tracking-app">Hari31416/typesafe-demo-finance-tracking-app</a></p>Classification & Routingtiered-thresholdsfailure-mode-knobhttps://jevbooks.com/en/p/mleyvaz__jev-typed-evaluation-collapsehttps://jevbooks.com/en/p/mleyvaz__jev-typed-evaluation-collapseCase study on Jev's typed evaluation collapse via six experimentsSun, 20 Sep 2026 08:49:16 GMT<p>Case study on Jev's typed evaluation collapse via six experiments</p><p>A field note and manuscript studying Jev (TypeSafe AI), a commercial 'System One model', through six experiments run via the Vercel AI Gateway, targeting the NCML journal.</p><p><a href="https://github.com/mleyvaz/jev-typed-evaluation-collapse">mleyvaz/jev-typed-evaluation-collapse</a></p>Evaluation & Benchmarkinghttps://jevbooks.com/en/p/weirdstuff-dev__typesafe-ai-rust-sdkhttps://jevbooks.com/en/p/weirdstuff-dev__typesafe-ai-rust-sdkUnofficial Rust SDK for TypeSafe AI, ported from JavaScript.Sun, 20 Sep 2026 08:49:03 GMT<p>Unofficial Rust SDK for TypeSafe AI, ported from JavaScript.</p><p>An unofficial Rust SDK for the TypeSafe AI API, ported from the official JavaScript SDK. It lets Rust developers call the TypeSafe AI System One model with typed questions and receive answers.</p><p><a href="https://github.com/weirdstuff-dev/typesafe-ai-rust-sdk">weirdstuff-dev/typesafe-ai-rust-sdk</a></p>Infra / SDKs / Integrationsflat-fanouthttps://jevbooks.com/en/p/stefw__lkcleanhttps://jevbooks.com/en/p/stefw__lkcleanChrome extension that filters LinkedIn feed noise with Jev and explains each verdict.Sun, 20 Sep 2026 08:49:03 GMT<p>Chrome extension that filters LinkedIn feed noise with Jev and explains each verdict.</p><p>A Manifest V3 TypeScript Chrome extension for LinkedIn users that collapses engagement bait, self-promotion, sponsored and off-topic posts into a one-line bar, and annotates kept posts with their scores.</p><p><a href="https://github.com/stefw/lkclean">stefw/lkclean</a></p>Content Moderationflat-fanoutcode-owned-policyhttps://jevbooks.com/en/p/miounet11__jevcodehttps://jevbooks.com/en/p/miounet11__jevcodeJev docs and best practices for engineering teamsSun, 20 Sep 2026 08:48:58 GMT<p>Jev docs and best practices for engineering teams</p><p>A Chinese-first documentation and solutions site for Jev, TypeSafe's System One decision model. It aggregates patterns, cases, and an automated publishing pipeline for engineers adopting Jev.</p><p><a href="https://github.com/miounet11/jevcode">miounet11/jevcode</a></p>Classification & Routinghttps://jevbooks.com/en/p/davidrydberg__git-judge-jevhttps://jevbooks.com/en/p/davidrydberg__git-judge-jevGitHub Action that judges agent-written PRs and posts one comment.Sun, 20 Sep 2026 08:48:47 GMT<p>GitHub Action that judges agent-written PRs and posts one comment.</p><p>A GitHub Action for pull requests written by coding agents. It reads the diff and posts a single comment saying what the PR really does, which hunks need a human, why, and what to verify.</p><p><a href="https://github.com/davidrydberg/git-judge-jev">davidrydberg/git-judge-jev</a></p>Verification & Guardrailsscreen-then-deephttps://jevbooks.com/en/p/tcsenpai__jevoraclehttps://jevbooks.com/en/p/tcsenpai__jevoracleTyped questions over your own state, not a chat.Sun, 20 Sep 2026 08:48:40 GMT<p>Typed questions over your own state, not a chat.</p><p>A small web console for asking Jev typed questions about pasted state. For developers who want to see the request shape and calibrated probabilities, not prose.</p><p><a href="https://github.com/tcsenpai/jevoracle">tcsenpai/jevoracle</a></p>Infra / SDKs / Integrationsflat-fanoutno-judgment-bandhttps://jevbooks.com/en/p/fatwang2__awesome-jevhttps://jevbooks.com/en/p/fatwang2__awesome-jevA source-backed directory of open-source projects built with TypeSafe Jev.Sun, 20 Sep 2026 08:48:38 GMT<p>A source-backed directory of open-source projects built with TypeSafe Jev.</p><p>A curated directory of 155 open-source projects built with TypeSafe Jev, spanning SDKs and integrations. It is aimed at developers who want to find or submit Jev-based tools, with submissions reviewed by Jev.</p><p><a href="https://github.com/fatwang2/awesome-jev">fatwang2/awesome-jev</a></p>Evaluation & Benchmarkinghttps://jevbooks.com/en/p/mrjev__awesome-jevhttps://jevbooks.com/en/p/mrjev__awesome-jevCurated list of projects that actually call Jev.Sun, 20 Sep 2026 08:48:34 GMT<p>Curated list of projects that actually call Jev.</p><p>A community-maintained, unofficial list of projects, integrations, and resources for Jev, TypeSafe AI's System One model, which answers typed questions about program state with calibrated confidence instead of generating text.</p><p><a href="https://github.com/MrJev/awesome-jev">MrJev/awesome-jev</a></p>Agent Decisionshttps://jevbooks.com/en/p/glud123__jev-assisthttps://jevbooks.com/en/p/glud123__jev-assistRank repo files by task relevance with Jev.Sun, 20 Sep 2026 08:48:30 GMT<p>Rank repo files by task relevance with Jev.</p><p>jev-assist is a CLI and agent skill for large repos. It ranks every tracked file against a one-line task, checks files against team conventions, and gates staged diffs for risk.</p><p><a href="https://github.com/glud123/jev-assist">glud123/jev-assist</a></p>Scoring & Rankingflat-fanouthttps://jevbooks.com/en/p/alexj11324__open-jev-approvalshttps://jevbooks.com/en/p/alexj11324__open-jev-approvalsBinary approval gate for Claude Code, reviewed by TypeSafe Jev.Sun, 20 Sep 2026 08:48:29 GMT<p>Binary approval gate for Claude Code, reviewed by TypeSafe Jev.</p><p>A Go approval gate for Claude Code. When the agent is about to run a tool, a hook intercepts the call, TypeSafe Jev reviews it, and a local policy returns allow or deny. Aimed at third-party API setups that cannot use the harness's own approval flow.</p><p><a href="https://github.com/alexj11324/open-jev-approvals">alexj11324/open-jev-approvals</a></p>Verification & Guardrailsflat-fanoutfailure-mode-knobhttps://jevbooks.com/en/p/chy4pro__jev-dev-kithttps://jevbooks.com/en/p/chy4pro__jev-dev-kitFramework for building agents on TypeSafe Jev.Sun, 20 Sep 2026 08:48:25 GMT<p>Framework for building agents on TypeSafe Jev.</p><p>A small framework for building agents on TypeSafe Jev. It turns candidates into valid Jev questions and answers into validated choices, with loop, cross-checks, fallbacks and traces built in. For developers building Jev-based agents.</p><p><a href="https://github.com/chy4pro/jev-dev-kit">chy4pro/jev-dev-kit</a></p>Agent Decisionsflat-fanoutcandidate-enumerationhttps://jevbooks.com/en/p/heman10x-ngu__openjev-verdict-2.0https://jevbooks.com/en/p/heman10x-ngu__openjev-verdict-2.0Calibrated 151M non-autoregressive decision engine for typed software workflows.Sun, 20 Sep 2026 08:48:19 GMT<p>Calibrated 151M non-autoregressive decision engine for typed software workflows.</p><p>openJev-verdict-2.0 is a 149.6M parameter decision model that evaluates typed decision schemas (Choice, Score, Noul) in non-autoregressive forward passes with calibrated confidence. It targets enterprise financial, security, customer support, and agent trace observability workflows.</p><p><a href="https://github.com/Heman10x-NGU/openJev-verdict-2.0">Heman10x-NGU/openJev-verdict-2.0</a></p>Evaluation & Benchmarkinghttps://jevbooks.com/en/p/24601__rh-guardhttps://jevbooks.com/en/p/24601__rh-guardReward-hack radar for coding agents: structural denies plus Jev sidecar.Sun, 20 Sep 2026 08:48:19 GMT<p>Reward-hack radar for coding agents: structural denies plus Jev sidecar.</p><p>rh-guard sits in coding-agent hooks (Claude Code, Cursor, Codex, Grok Build, Pi, Amp, Prime Agent, DeepSeek Harness) and blocks reward-hacking tool use—tampering with graders, hidden tests, or the eval process—while steering toward checks the agent cannot game.</p><p><a href="https://github.com/24601/rh-guard">24601/rh-guard</a></p>Verification & Guardrailshttps://jevbooks.com/en/p/everyinfra__jev-radarhttps://jevbooks.com/en/p/everyinfra__jev-radarIndependent tracker of the Jev ecosystem with verified cases.Sun, 20 Sep 2026 08:48:12 GMT<p>Independent tracker of the Jev ecosystem with verified cases.</p><p>Jev Radar is an independent field report and live monitor of the Jev (TypeSafe AI System One) ecosystem, cataloging 108 confidence-graded projects for developers and researchers.</p><p><a href="https://github.com/everyinfra/jev-radar">everyinfra/jev-radar</a></p>Calibration & Researchflat-fanouthttps://jevbooks.com/en/p/nozomi-koborinai__jev-spechttps://jevbooks.com/en/p/nozomi-koborinai__jev-specSemantic verification engine for AI code and specs, powered by Jev.Sun, 20 Sep 2026 08:48:12 GMT<p>Semantic verification engine for AI code and specs, powered by Jev.</p><p>jev-spec is a Specification-Driven Development verification tool for teams using AI coding assistants. It checks whether code satisfies written requirements, catching semantic gaps that structural linters miss.</p><p><a href="https://github.com/nozomi-koborinai/jev-spec">nozomi-koborinai/jev-spec</a></p>Verification & Guardrailshttps://jevbooks.com/en/p/perrylink__jevcorehttps://jevbooks.com/en/p/perrylink__jevcoreTypeSafe Jev for DeepSeek Harness and MCP hosts.Sun, 20 Sep 2026 08:48:03 GMT<p>TypeSafe Jev for DeepSeek Harness and MCP hosts.</p><p>jevcore gives agents a typed judgment surface: Jev answers noul, choice, and score questions with calibrated probabilities instead of prose. It ships as a plain Node library, a DeepSeek Harness plugin, and an MCP stdio server.</p><p><a href="https://github.com/PerryLink/jevcore">PerryLink/jevcore</a></p>Agent Decisionshttps://jevbooks.com/en/p/cyvid7-darus10__docfrictionhttps://jevbooks.com/en/p/cyvid7-darus10__docfrictionAutomated friction logs for documentation, scored section by section with TypeSafe Jev.Sun, 20 Sep 2026 08:48:03 GMT<p>Automated friction logs for documentation, scored section by section with TypeSafe Jev.</p><p>docfriction generates friction logs for online documentation pages, walking through each section and asking TypeSafe Jev a fixed set of typed questions. It is for developer advocates and writers who want to find where readers get stuck.</p><p><a href="https://github.com/Cyvid7-Darus10/docfriction">Cyvid7-Darus10/docfriction</a></p>Evaluation & Benchmarkinghttps://jevbooks.com/en/p/yzfly__awesome-jev-zhhttps://jevbooks.com/en/p/yzfly__awesome-jev-zhCurated Chinese list of Jev resources, SDKs, apps, and guides.Sun, 20 Sep 2026 08:47:55 GMT<p>Curated Chinese list of Jev resources, SDKs, apps, and guides.</p><p>A Chinese-language awesome list for Jev, the TypeSafe System One model. It collects official docs, SDKs, popular apps, agent tools, open-source reimplementations, and independent evaluations, plus Chinese guides and a daily auto-updated GitHub trending list.</p><p><a href="https://github.com/yzfly/awesome-jev-zh">yzfly/awesome-jev-zh</a></p>Classification & Routingflat-fanouthttps://jevbooks.com/en/p/jaykang-heo__jev-axihttps://jevbooks.com/en/p/jaykang-heo__jev-axiAgent-ergonomic CLI for TypeSafe Jev typed judgmentsSun, 20 Sep 2026 08:47:52 GMT<p>Agent-ergonomic CLI for TypeSafe Jev typed judgments</p><p>A TypeScript CLI that asks TypeSafe's Jev for typed judgments about a text: pick one option from a closed set, give a yes/no probability, or score against a rubric. Built for agents and developers who threshold on numbers.</p><p><a href="https://github.com/jaykang-heo/jev-axi">jaykang-heo/jev-axi</a></p>Agent Decisionsflat-fanouthttps://jevbooks.com/en/p/hoangngochuong24947-gif__jev-figure-routerhttps://jevbooks.com/en/p/hoangngochuong24947-gif__jev-figure-routerUniversal figure and diagram router for AI agents, powered by TypeSafe Jev.Sun, 20 Sep 2026 08:47:41 GMT<p>Universal figure and diagram router for AI agents, powered by TypeSafe Jev.</p><p>A router that maps an AI agent's visualization request to one of six rendering branches (scientific plots, architecture diagrams, slides, AI-generated visuals, materials computation, chart MCP), returning a target tool and output format.</p><p><a href="https://github.com/hoangngochuong24947-gif/jev-figure-router">hoangngochuong24947-gif/jev-figure-router</a></p>Classification & Routingscreen-then-deephttps://jevbooks.com/en/p/zhuyansen__x-reply-filterhttps://jevbooks.com/en/p/zhuyansen__x-reply-filterChrome extension collapses low-quality X replies using local rules and TypeSafe Jev.Sun, 20 Sep 2026 08:47:40 GMT<p>Chrome extension collapses low-quality X replies using local rules and TypeSafe Jev.</p><p>A Manifest V3 Chrome extension for X (x.com) tweet detail pages that collapses spam, bait, off-topic and AI-filler replies. Two-stage: free local rules in rules.js, then Jev for the rest. Users see hidden replies as collapsible grey bars and learn from manual corrections.</p><p><a href="https://github.com/zhuyansen/x-reply-filter">zhuyansen/x-reply-filter</a></p>Content Moderationflat-fanouthttps://jevbooks.com/en/p/nixz0824__rag-jevhttps://jevbooks.com/en/p/nixz0824__rag-jevLocal RAG over Chinese LoL patch notes, with Jev for reranking and self-check.Sun, 20 Sep 2026 08:47:34 GMT<p>Local RAG over Chinese LoL patch notes, with Jev for reranking and self-check.</p><p>A local RAG Q&amp;A system for the Chinese server's League of Legends patch notes. Numbers in answers come only from structured extraction of official announcements, never from a generative model. Built for users asking about champion, item, and system changes across 30 versions.</p><p><a href="https://github.com/Nixz0824/rag-jev">Nixz0824/rag-jev</a></p>Scoring & Rankingcontract-validationhttps://jevbooks.com/en/p/typesafeai__clarity-judgehttps://jevbooks.com/en/p/typesafeai__clarity-judgeMulti-axis writing quality checker powered by TypeSafe AI's Jev model.Sun, 20 Sep 2026 08:47:34 GMT<p>Multi-axis writing quality checker powered by TypeSafe AI's Jev model.</p><p>A community-built TypeScript app that evaluates supplied writing against separate named checks—hedging, em dashes, clarity, filler, tone, passive voice, and actionability—showing each verdict, confidence, and supporting sentence instead of one opaque score.</p><p><a href="https://github.com/TypeSafeAI/clarity-judge">TypeSafeAI/clarity-judge</a></p>Evaluation & Benchmarkinghttps://jevbooks.com/en/p/himomohi__aside-jevhttps://jevbooks.com/en/p/himomohi__aside-jevAside agents decide with TypeSafe Jev (System One: Choice/Score/Noul).Sun, 20 Sep 2026 08:47:32 GMT<p>Aside agents decide with TypeSafe Jev (System One: Choice/Score/Noul).</p><p>Aside Jev connects the Aside browser REPL with TypeSafe Jev. It observes the page, compacts context, and lets Jev pick the next action from user-defined candidates. For developers automating browser tasks.</p><p><a href="https://github.com/himomohi/aside-jev">himomohi/aside-jev</a></p>Agent Decisionshttps://jevbooks.com/en/p/rudyjunyu__jev-mcphttps://jevbooks.com/en/p/rudyjunyu__jev-mcpGo MCP gateway to TypeSafe's Jev modelSun, 20 Sep 2026 08:47:25 GMT<p>Go MCP gateway to TypeSafe's Jev model</p><p>A small Go gateway that exposes TypeSafe's Jev model through the Model Context Protocol, forwarding each caller's TypeSafe API key to the System One API for MCP clients such as Codex and Claude Code.</p><p><a href="https://github.com/RudyJunyu/Jev-MCP">RudyJunyu/Jev-MCP</a></p>Infra / SDKs / Integrationsflat-fanouthttps://jevbooks.com/en/p/hoangngochuong24947-gif__jev-patent-disclosurehttps://jevbooks.com/en/p/hoangngochuong24947-gif__jev-patent-disclosureJev gates patent drafting: 300ms fact check before LLM writing.Sun, 20 Sep 2026 08:47:21 GMT<p>Jev gates patent drafting: 300ms fact check before LLM writing.</p><p>A Chinese patent skill suite for R&amp;D teams and patent engineers: mining patent points, scoring fact completeness, judging legal patent type, and generating disclosure/application documents, OA replies and Obsidian patent maps.</p><p><a href="https://github.com/hoangngochuong24947-gif/jev-patent-disclosure">hoangngochuong24947-gif/jev-patent-disclosure</a></p>Verification & Guardrailsscreen-then-deephttps://jevbooks.com/en/p/rickysullivan__magic-jev-ballhttps://jevbooks.com/en/p/rickysullivan__magic-jev-ballSpeech-only Magic 8 Ball with Jev making bounded decisions.Sun, 20 Sep 2026 08:47:17 GMT<p>Speech-only Magic 8 Ball with Jev making bounded decisions.</p><p>A phone-first, speech-only Magic 8 Ball experience for players. Visitors speak a question or assertion, shake or tap, and silently watch a response resolve through a dark-blue liquid window.</p><p><a href="https://github.com/rickysullivan/magic-jev-ball">rickysullivan/magic-jev-ball</a></p>Game & Simulationhttps://jevbooks.com/en/p/cobanov__awesome-jevhttps://jevbooks.com/en/p/cobanov__awesome-jevCurated list of projects built with Jev, TypeSafe AI's System One model.Sun, 20 Sep 2026 08:47:14 GMT<p>Curated list of projects built with Jev, TypeSafe AI's System One model.</p><p>A source-backed catalog of projects using Jev, TypeSafe AI's System One model for typed probabilistic decisions. It targets developers seeking concrete Jev integrations, SDKs, and research, favoring public code and reproducible evidence.</p><p><a href="https://github.com/cobanov/awesome-jev">cobanov/awesome-jev</a></p>Otherhttps://jevbooks.com/en/p/jyeswak__jev_playgroundhttps://jevbooks.com/en/p/jyeswak__jev_playgroundMeasure Jev's real capabilities before building on it.Sun, 20 Sep 2026 08:46:25 GMT<p>Measure Jev's real capabilities before building on it.</p><p>A playground that tests Jev against real data, documenting graded findings, ruled-out candidates, and recipes with stop-conditions. For developers evaluating Jev's cost-benefit before integration.</p><p><a href="https://github.com/JYeswak/jev_playground">JYeswak/jev_playground</a></p>Evaluation & Benchmarkingmeasured-calibrationhttps://jevbooks.com/en/p/radrebelsam__awesome-jevhttps://jevbooks.com/en/p/radrebelsam__awesome-jevDirectory of projects built on Jev, TypeSafe AI's System One model.Sun, 20 Sep 2026 08:46:22 GMT<p>Directory of projects built on Jev, TypeSafe AI's System One model.</p><p>A crawler-maintained directory of open-source projects built on Jev, TypeSafe AI's System One model. Daily GitHub and npm sweep, human-merged. For developers exploring Jev-based tools, SDKs, integrations and examples.</p><p><a href="https://github.com/RadRebelSam/awesome-jev">RadRebelSam/awesome-jev</a></p>Infra / SDKs / Integrationsflat-fanouthttps://jevbooks.com/en/p/kofanlabs__jev-browser-chromehttps://jevbooks.com/en/p/kofanlabs__jev-browser-chromeWindows Chrome/Edge browser agent using Jev for ultrafast tab control.Sun, 20 Sep 2026 08:46:20 GMT<p>Windows Chrome/Edge browser agent using Jev for ultrafast tab control.</p><p>A Windows adaptation of Jev Ultrafast for Chrome and Edge, giving agents like Codex and Grok browser access through a local extension bridge and MCP. It controls existing signed-in tabs without Chrome remote debugging or profile copying.</p><p><a href="https://github.com/kofanlabs/jev-browser-chrome">kofanlabs/jev-browser-chrome</a></p>Agent Decisionsspeculative-headscandidate-enumerationhttps://jevbooks.com/en/p/7starsseeker__dsh-jev-guardhttps://jevbooks.com/en/p/7starsseeker__dsh-jev-guardPre-execution safety valve for DeepSeek Harness shell commands.Sun, 20 Sep 2026 08:46:16 GMT<p>Pre-execution safety valve for DeepSeek Harness shell commands.</p><p>A DSH plugin that intercepts bash/pwsh tool calls before they run, judging whether a command will irreversibly delete or overwrite real data. Built for DSH users who want a hard interception rather than a model nudge.</p><p><a href="https://github.com/7starsseeker/dsh-jev-guard">7starsseeker/dsh-jev-guard</a></p>Verification & Guardrailshttps://jevbooks.com/en/p/armansra-hub__stanleyhttps://jevbooks.com/en/p/armansra-hub__stanleyTerritory intelligence for a NetSuite AE, with Jev and Claude.Sun, 20 Sep 2026 08:46:16 GMT<p>Territory intelligence for a NetSuite AE, with Jev and Claude.</p><p>Stanley is a single-user prospecting and workflow assistant for a NetSuite account executive. It combines a hosted Next.js app with local, supervised sales workflows for TAM/TAL management, account research, missions, and pipeline tracking.</p><p><a href="https://github.com/armansra-hub/stanley">armansra-hub/stanley</a></p>Agent Decisionshttps://jevbooks.com/en/p/kofanlabs__typesafe-computer-use-windowshttps://jevbooks.com/en/p/kofanlabs__typesafe-computer-use-windowsWindows desktop automation where Jev picks each operation and target.Sun, 20 Sep 2026 08:46:15 GMT<p>Windows desktop automation where Jev picks each operation and target.</p><p>A Windows port of TypeSafe Computer Use for agents like Codex and Grok. Given a goal and an open Windows application, it automates the task using local OCR, UI Automation, and MCP host integration.</p><p><a href="https://github.com/kofanlabs/typesafe-computer-use-windows">kofanlabs/typesafe-computer-use-windows</a></p>Agent Decisionshttps://jevbooks.com/en/p/elcronos__jev-vs-open-decision-modelshttps://jevbooks.com/en/p/elcronos__jev-vs-open-decision-modelsZero-shot benchmark: Jev vs open decision models on classification.Sun, 20 Sep 2026 08:46:05 GMT<p>Zero-shot benchmark: Jev vs open decision models on classification.</p><p>A frozen-protocol benchmark comparing Jev 1.13, PrismNLI-0.4B and Laya on four public English classification datasets, with supervised reference models and a label-efficiency study for practitioners deciding between zero-shot and labelling.</p><p><a href="https://github.com/elcronos/jev-vs-open-decision-models">elcronos/jev-vs-open-decision-models</a></p>Evaluation & Benchmarkinghttps://jevbooks.com/en/p/abhitsian__seekhttps://jevbooks.com/en/p/abhitsian__seekMenu-bar Mac search that finds files, apps, tabs and people by description.Sun, 20 Sep 2026 08:45:58 GMT<p>Menu-bar Mac search that finds files, apps, tabs and people by description.</p><p>Seek is a menu-bar Mac app for finding files by describing them. Press ⌃⌥F, type a description, and it shows matching files, apps, System Settings pages, Chrome tabs, Teams chats and Claude Code sessions.</p><p><a href="https://github.com/abhitsian/seek">abhitsian/seek</a></p>Scoring & Rankinghttps://jevbooks.com/en/p/waterme7on__jev-paper-traderhttps://jevbooks.com/en/p/waterme7on__jev-paper-traderJev as decision engine for BTC/ETH paper trading, evaluated every 5 seconds.Sun, 20 Sep 2026 08:45:52 GMT<p>Jev as decision engine for BTC/ETH paper trading, evaluated every 5 seconds.</p><p>A browser-based BTC/ETH paper trading desk. Every 5 seconds it evaluates the market, draws buy/sell points and decision history. No exchange connection, no real orders, no funds; account and positions live in localStorage.</p><p><a href="https://github.com/waterme7on/jev-paper-trader">waterme7on/jev-paper-trader</a></p>Finance & Tradinghttps://jevbooks.com/en/p/majiayu000__awesome-jevhttps://jevbooks.com/en/p/majiayu000__awesome-jevCurated bilingual list of Jev projects, SDKs, tutorials, and evaluations.Sun, 20 Sep 2026 08:45:45 GMT<p>Curated bilingual list of Jev projects, SDKs, tutorials, and evaluations.</p><p>A community-maintained awesome list of open-source projects, SDKs, tutorials, and evaluations for Jev, TypeSafe's System One model, organized by category with English and Simplified Chinese resources.</p><p><a href="https://github.com/majiayu000/awesome-jev">majiayu000/awesome-jev</a></p>Otherhttps://jevbooks.com/en/p/dtheofr__typesafe-jev-rubyhttps://jevbooks.com/en/p/dtheofr__typesafe-jev-rubyRuby client for Jev: typed, probabilistic answers from application state.Sun, 20 Sep 2026 08:45:42 GMT<p>Ruby client for Jev: typed, probabilistic answers from application state.</p><p>A Ruby gem that wraps Jev, TypeSafe's System One model, letting Ruby and Rails developers ask natural-language questions about application state and receive typed, probabilistic answers. Zero runtime dependencies.</p><p><a href="https://github.com/dtheofr/typesafe-jev-ruby">dtheofr/typesafe-jev-ruby</a></p>Infra / SDKs / Integrationsflat-fanouthttps://jevbooks.com/en/p/ygivenx__jev-tryhttps://jevbooks.com/en/p/ygivenx__jev-tryReads clinical note, nudges for orders committed but not charted.Sun, 20 Sep 2026 08:45:38 GMT<p>Reads clinical note, nudges for orders committed but not charted.</p><p>A single-page app that reads a clinical note and tells you what the note commits to that the chart does not yet carry out. Built on TypeSafe's jev model. For demonstration only, not for clinical use.</p><p><a href="https://github.com/ygivenx/jev-try">ygivenx/jev-try</a></p>Verification & Guardrailstiered-thresholdsno-judgment-bandcode-owned-policyhttps://jevbooks.com/en/p/ycs77__jev-girlfriend-analysishttps://jevbooks.com/en/p/ycs77__jev-girlfriend-analysisGuess your girlfriend's real thoughts with Jev.Sun, 20 Sep 2026 08:45:27 GMT<p>Guess your girlfriend's real thoughts with Jev.</p><p>An experimental TypeScript project for people in or after a relationship. It reads a message plus optional context and outputs a possible mood state and one concrete action, with separate scripts for a current girlfriend and an ex-girlfriend.</p><p><a href="https://github.com/ycs77/jev-girlfriend-analysis">ycs77/jev-girlfriend-analysis</a></p>Agent Decisionshttps://jevbooks.com/en/p/heliowap__diff-risk-sentinelhttps://jevbooks.com/en/p/heliowap__diff-risk-sentinelRisk triage for large git diffs and repository-wide dead-code scan.Sun, 20 Sep 2026 08:45:22 GMT<p>Risk triage for large git diffs and repository-wide dead-code scan.</p><p>Diff Risk Sentinel ranks methods touched by a git diff by structural risk (CRAP) and optional semantic risk, helping reviewers prioritize. It also scans a whole repository for dead code in Python and JS/TS.</p><p><a href="https://github.com/heliowap/diff-risk-sentinel">heliowap/diff-risk-sentinel</a></p>Classification & Routinghttps://jevbooks.com/en/p/omkarghugarkar007__actiongate-jevhttps://jevbooks.com/en/p/omkarghugarkar007__actiongate-jevRuntime authorization gateway for AI agent tool calls using Jev.Sun, 20 Sep 2026 08:45:19 GMT<p>Runtime authorization gateway for AI agent tool calls using Jev.</p><p>ActionGate is a runtime security gateway that evaluates proposed tool calls with deterministic policy and Jev, then issues single-use permits bound to the exact action. It is for developers building AI agents that need to authorize tool calls.</p><p><a href="https://github.com/omkarghugarkar007/actiongate-jev">omkarghugarkar007/actiongate-jev</a></p>Verification & Guardrailshttps://jevbooks.com/en/p/thomasbrueggemann__jeffreyhttps://jevbooks.com/en/p/thomasbrueggemann__jeffreyCoding-agent CLI where Jev decides and an LLM executes.Sun, 20 Sep 2026 08:45:13 GMT<p>Coding-agent CLI where Jev decides and an LLM executes.</p><p>jeffrey is a coding-agent CLI for developers. A decision model picks the next tool and scores progress, risk, and goal completion; an OpenAI-compatible LLM fills in the tool arguments and does the actual work.</p><p><a href="https://github.com/thomasbrueggemann/jeffrey">thomasbrueggemann/jeffrey</a></p>Agent Decisionshttps://jevbooks.com/en/p/yo4e__jevpiphttps://jevbooks.com/en/p/yo4e__jevpipLocal GMO FX/BTC market research terminal with TypeSafe Jev paper-trading integration.Sun, 20 Sep 2026 08:45:12 GMT<p>Local GMO FX/BTC market research terminal with TypeSafe Jev paper-trading integration.</p><p>JevPip is a local market research terminal and study app for GMO FX (12 JPY pairs) and BTC/JPY, bundling live charts, raw tick collection, paper trading, backtests, and a safety supervisor into one local UI. It does not place real trades.</p><p><a href="https://github.com/yo4e/JevPip">yo4e/JevPip</a></p>Finance & Tradinghttps://jevbooks.com/en/p/luizribeiro__jevrshttps://jevbooks.com/en/p/luizribeiro__jevrsAsync Rust client for TypeSafe AI's Jev with typed questions and WASI transports.Sun, 20 Sep 2026 08:45:10 GMT<p>Async Rust client for TypeSafe AI's Jev with typed questions and WASI transports.</p><p>jevrs is an async Rust client for TypeSafe AI's Jev (System One) API, giving Rust developers typed question sets and answers, a runtime builder, configurable retries, and native or WASI transports over a portable sans-I/O core.</p><p><a href="https://github.com/luizribeiro/jevrs">luizribeiro/jevrs</a></p>Infra / SDKs / Integrationshttps://jevbooks.com/en/p/emreozyoruk__hushhttps://jevbooks.com/en/p/emreozyoruk__hushIssue triage that stays quiet when it isn't sure.Sun, 20 Sep 2026 08:45:05 GMT<p>Issue triage that stays quiet when it isn't sure.</p><p>A GitHub Action that labels new issues and pull requests only when confident, otherwise abstains. For maintainers who want calibrated triage without noisy false positives.</p><p><a href="https://github.com/emreozyoruk/hush">emreozyoruk/hush</a></p>Classification & Routingtiered-thresholdshttps://jevbooks.com/en/p/jaswanthsanjay88__revhttps://jevbooks.com/en/p/jaswanthsanjay88__revFast, prefill-only decision model with typed questions and calibrated probabilities.Sun, 20 Sep 2026 08:44:58 GMT<p>Fast, prefill-only decision model with typed questions and calibrated probabilities.</p><p>rev is a decision model built on a causal LM backbone (Qwen/Qwen2.5-0.5B to 8B) with a LoRA adapter and bilinear pointer readout head. It reads a document once and evaluates multiple typed questions in parallel in a single prefill forward pass with zero autoregressive text generation. Weights are on Hugging Face Hub.</p><p><a href="https://github.com/jaswanthsanjay88/rev">jaswanthsanjay88/rev</a></p>Classification & Routinghttps://jevbooks.com/en/p/dbreunig__building-with-jev-skillhttps://jevbooks.com/en/p/dbreunig__building-with-jev-skillAgent skill for writing and improving programs that call Jev.Sun, 20 Sep 2026 08:44:32 GMT<p>Agent skill for writing and improving programs that call Jev.</p><p>An agent skill that helps developers write and improve programs calling Jev, TypeSafe's System One model. It covers question design, state structure, answer composition, confidence thresholds, and debugging. Targets jev-1.13.</p><p><a href="https://github.com/dbreunig/building-with-jev-skill">dbreunig/building-with-jev-skill</a></p>Evaluation & Benchmarkinghttps://jevbooks.com/en/p/zhuyansen__jev-search-rerank-evalhttps://jevbooks.com/en/p/zhuyansen__jev-search-rerank-evalDoes a TypeSafe Jev rerank beat embedding search?Sun, 20 Sep 2026 08:44:28 GMT<p>Does a TypeSafe Jev rerank beat embedding search?</p><p>A graded relevance evaluation over the Agent Skills Hub catalog (33,047 skills, MCP servers, coding-agent tools) with 164 Chinese/English/mixed queries and 9,831 labelled pairs, comparing Jev reranking against lexical, embedding, and fusion baselines.</p><p><a href="https://github.com/zhuyansen/jev-search-rerank-eval">zhuyansen/jev-search-rerank-eval</a></p>Scoring & Rankinghttps://jevbooks.com/en/p/mahlernim__jev-korean-benchmarkhttps://jevbooks.com/en/p/mahlernim__jev-korean-benchmarkReproducible early-access check of Jev on Korean and medical text.Sun, 20 Sep 2026 08:44:19 GMT<p>Reproducible early-access check of Jev on Korean and medical text.</p><p>A small frozen study that asks 100 questions per cell from four public test sets (Belebele, PAWS-X, MedQA, KorMedMCQA) in Korean and English, recording every Jev response. It is a sample check, not a benchmark.</p><p><a href="https://github.com/mahlernim/jev-korean-benchmark">mahlernim/jev-korean-benchmark</a></p>Evaluation & Benchmarkinghttps://jevbooks.com/en/p/vinilana__jev-eval-agenthttps://jevbooks.com/en/p/vinilana__jev-eval-agentCompares LLM tool selection with Jev routing in a 100-tool agent.Sun, 20 Sep 2026 08:44:18 GMT<p>Compares LLM tool selection with Jev routing in a 100-tool agent.</p><p>A personal-assistant agent built with eve (Vercel), served through OpenRouter, containing 100 mocked tools. It benchmarks how many steps the agent needs when the LLM picks the tool itself versus when Jev picks it.</p><p><a href="https://github.com/vinilana/jev-eval-agent">vinilana/jev-eval-agent</a></p>Agent Decisionsflat-fanoutcross-checkhttps://jevbooks.com/en/p/4esv__jev-evalhttps://jevbooks.com/en/p/4esv__jev-evalBenchmark TypeSafe Jev against any OpenRouter model on labelled classification data.Sun, 20 Sep 2026 08:44:12 GMT<p>Benchmark TypeSafe Jev against any OpenRouter model on labelled classification data.</p><p>A Python benchmark that compares TypeSafe Jev with any OpenRouter model on labelled classification tasks, reporting accuracy, calibration, latency, cost and determinism. It runs on three included public tasks or your own data in data/.</p><p><a href="https://github.com/4esv/jev-eval">4esv/jev-eval</a></p>Evaluation & Benchmarkingmeasured-calibrationhttps://jevbooks.com/en/p/baibizhe__jev-decision-benchmarkshttps://jevbooks.com/en/p/baibizhe__jev-decision-benchmarksIndependent JEV benchmarks on MetaTool, When2Call, and BFCL V4.Sun, 20 Sep 2026 08:44:09 GMT<p>Independent JEV benchmarks on MetaTool, When2Call, and BFCL V4.</p><p>Evaluation reports of JEV on three agent decision benchmarks, comparing against Claude, Qwen, GPT and other models. For readers assessing JEV tool selection and abstention behavior.</p><p><a href="https://github.com/baibizhe/jev-decision-benchmarks">baibizhe/jev-decision-benchmarks</a></p>Evaluation & Benchmarkinghttps://jevbooks.com/en/p/abdelstark__jev-benchmarkshttps://jevbooks.com/en/p/abdelstark__jev-benchmarksProbability-aware evaluation for typed decision models: calibration, selective risk, latency.Sun, 20 Sep 2026 08:44:06 GMT<p>Probability-aware evaluation for typed decision models: calibration, selective risk, latency.</p><p>jev-benchmarks is a reproducible evaluation harness for typed decision models. It measures calibration, selective automation at a fixed error budget, latency, and integrity, and ships a pilot comparing TypeSafe Jev with GLiNER2.5 on zero-shot single-label text classification.</p><p><a href="https://github.com/AbdelStark/jev-benchmarks">AbdelStark/jev-benchmarks</a></p>Evaluation & Benchmarkingmeasured-calibrationhttps://jevbooks.com/en/p/rinnecoder__jev-behavior-studyhttps://jevbooks.com/en/p/rinnecoder__jev-behavior-studyIndependent behavior study of Jev 1.13.0: findings and failures.Sun, 20 Sep 2026 08:44:02 GMT<p>Independent behavior study of Jev 1.13.0: findings and failures.</p><p>A public field guide to Jev 1.13.0's framing sensitivity and failures, based on 11,621 text-study requests across synthetic tasks. For AI researchers and practitioners choosing how to interact with Jev.</p><p><a href="https://github.com/RINNECODER/jev-behavior-study">RINNECODER/jev-behavior-study</a></p>Calibration & Researchhttps://jevbooks.com/en/p/heman10x-ngu__verdict-open-jevhttps://jevbooks.com/en/p/heman10x-ngu__verdict-open-jevNon-autoregressive decision engine with calibrated uncertainty and WebGPU playground.Sun, 20 Sep 2026 08:43:58 GMT<p>Non-autoregressive decision engine with calibrated uncertainty and WebGPU playground.</p><p>OpenJev (Verdict) is an open-source post-trained foundational decision model for structured software workflows, inspired by TypeSafe AI's Jev and RLCD. It evaluates typed questions in a single forward pass, returning choices, scores, and probabilities with calibrated confidence.</p><p><a href="https://github.com/Heman10x-NGU/Verdict-open-jev">Heman10x-NGU/Verdict-open-jev</a></p>Agent Decisionsflat-fanouthttps://jevbooks.com/en/p/featherless-ai__simple-jevhttps://jevbooks.com/en/p/featherless-ai__simple-jevTurn open models into typed classifiers without a trained head.Sun, 20 Sep 2026 08:43:54 GMT<p>Turn open models into typed classifiers without a trained head.</p><p>Simple Jev lets you use compatible open models from Hugging Face for structured classification and scoring, without training a separate classifier head. It is for developers who want typed decisions from a model's next-token logits.</p><p><a href="https://github.com/featherless-ai/simple-jev">featherless-ai/simple-jev</a></p>Classification & Routingflat-fanouthttps://jevbooks.com/en/p/kshetrajna12__reflexhttps://jevbooks.com/en/p/kshetrajna12__reflexA tiny decision model: state plus typed questions, calibrated probabilities.Sun, 20 Sep 2026 08:43:46 GMT<p>A tiny decision model: state plus typed questions, calibrated probabilities.</p><p>An open re-creation of Jev, TypeSafe's System One model, built on Qwen3.5-4B. It answers typed questions about a state (text or image) with calibrated probabilities, for developers who need fast structured judgments instead of prose.</p><p><a href="https://github.com/kshetrajna12/reflex">kshetrajna12/reflex</a></p>Classification & Routingflat-fanoutmeasured-calibrationhttps://jevbooks.com/en/p/ekzhang__openjev-sglanghttps://jevbooks.com/en/p/ekzhang__openjev-sglangJev-compatible HTTP API served by Qwen3.6-35B-A3B on SGLang.Sun, 20 Sep 2026 08:43:43 GMT<p>Jev-compatible HTTP API served by Qwen3.6-35B-A3B on SGLang.</p><p>A server implementing the TypeSafe/Jev HTTP API with Qwen3.6-35B-A3B on SGLang, using prefill-only one-token inference. It targets developers who want a self-hosted, Jev-compatible evaluation endpoint for Noul, Choice, and Score questions.</p><p><a href="https://github.com/ekzhang/openjev-sglang">ekzhang/openjev-sglang</a></p>Infra / SDKs / Integrationsflat-fanouthttps://jevbooks.com/en/p/razorback16__openjevhttps://jevbooks.com/en/p/razorback16__openjevOpen, Jev-compatible System One decision server on DiffusionGemma.Sun, 20 Sep 2026 08:43:41 GMT<p>Open, Jev-compatible System One decision server on DiffusionGemma.</p><p>OpenJev is an open-source System One decision server. Send a state and typed questions (yes/no, choice, score); it returns calibrated probabilities and confidence in tens of milliseconds. It runs DiffusionGemma 26B-A4B via vLLM on NVIDIA GPU or MLX on Apple silicon.</p><p><a href="https://github.com/razorback16/openjev">razorback16/openjev</a></p>Infra / SDKs / Integrationsflat-fanouthttps://jevbooks.com/en/p/bnsd55__jevmlxhttps://jevbooks.com/en/p/bnsd55__jevmlxParallel constrained decisions for MLX models on Apple Silicon.Sun, 20 Sep 2026 08:43:37 GMT<p>Parallel constrained decisions for MLX models on Apple Silicon.</p><p>jevmlx turns a schema of fields (booleans, enums, multi-selects) and a context string into a single batched forward pass on a local Apple Silicon model, returning schema-valid JSON with per-field probabilities.</p><p><a href="https://github.com/bnsd55/jevmlx">bnsd55/jevmlx</a></p>Classification & Routingflat-fanouthttps://jevbooks.com/en/p/hr98w__jev-visualhttps://jevbooks.com/en/p/hr98w__jev-visualEducational MLX vision-language inference with shared image context and direct candidate scoring.Sun, 20 Sep 2026 08:43:33 GMT<p>Educational MLX vision-language inference with shared image context and direct candidate scoring.</p><p>A small runnable Python project for learning vision-language model inference on Apple Silicon. It uses Qwen3.5-0.8B with MLX to answer multiple questions about one image, with a local browser UI, CLI and HTTP API.</p><p><a href="https://github.com/hr98w/jev-visual">hr98w/jev-visual</a></p>Scoring & Rankingflat-fanouthttps://jevbooks.com/en/p/rafalwilinski__vibecheckhttps://jevbooks.com/en/p/rafalwilinski__vibecheckChrome extension: vibe-check your X posts with TypeSafe's Jev before you hit PostSun, 20 Sep 2026 08:43:27 GMT<p>Chrome extension: vibe-check your X posts with TypeSafe's Jev before you hit Post</p><p>A Chrome and Firefox extension that scores draft X posts and reply context before posting, rendering a scorecard with virality, funniness, informativeness, clarity, insult level, ragebait, cringe, &quot;sounds AI-written&quot;, regret risk, typo detection, and likely crowd reaction, plus an overall verdict.</p><p><a href="https://github.com/RafalWilinski/vibecheck">RafalWilinski/vibecheck</a></p>Scoring & Rankinghttps://jevbooks.com/en/p/trungdq88__youtube-sponsor-detectionhttps://jevbooks.com/en/p/trungdq88__youtube-sponsor-detectionSkip YouTube sponsor reads with Jev-typed questionsSun, 20 Sep 2026 08:43:25 GMT<p>Skip YouTube sponsor reads with Jev-typed questions</p><p>A Chrome extension and small web app that find sponsor reads in YouTube videos and skip them. Built for viewers who want sponsor segments detected from transcripts or live audio.</p><p><a href="https://github.com/trungdq88/youtube-sponsor-detection">trungdq88/youtube-sponsor-detection</a></p>Classification & Routinghttps://jevbooks.com/en/p/harshil1712__slidepilothttps://jevbooks.com/en/p/harshil1712__slidepilotVoice-driven semantic auto-advance for Slidev, powered by Cloudflare Agents and Jev.Sun, 20 Sep 2026 08:43:22 GMT<p>Voice-driven semantic auto-advance for Slidev, powered by Cloudflare Agents and Jev.</p><p>An experimental Slidev addon that listens to a presenter's speech and automatically advances slides when the current slide is semantically complete, with deterministic safety checks and manual navigation always available.</p><p><a href="https://github.com/harshil1712/slidepilot">harshil1712/slidepilot</a></p>Verification & Guardrailsflat-fanouthttps://jevbooks.com/en/p/cocktailpeanut__jevthovenhttps://jevbooks.com/en/p/cocktailpeanut__jevthovenPrompt-first symbolic-music studio where Jev decides every musical unit.Sun, 20 Sep 2026 08:43:19 GMT<p>Prompt-first symbolic-music studio where Jev decides every musical unit.</p><p>Jevthoven turns a one-sentence music description into an editable multitrack composition. It is for users who want to play, reshape, and export symbolic music and MIDI, with every note traceable to a persisted Jev decision.</p><p><a href="https://github.com/cocktailpeanut/jevthoven">cocktailpeanut/jevthoven</a></p>Game & Simulationhttps://jevbooks.com/en/p/chetaslua__jevmeterhttps://jevbooks.com/en/p/chetaslua__jevmeterLive Jev BS meter over any video, rendered as a 16:9 edit.Sun, 20 Sep 2026 08:43:15 GMT<p>Live Jev BS meter over any video, rendered as a 16:9 edit.</p><p>jevmeter scores every sentence in a video with Jev and renders the result as an overlay edit. It is for people who want a shareable, sentence-by-sentence meter of debates, earnings calls, podcasts or pitches.</p><p><a href="https://github.com/ChetasLua/jevmeter">ChetasLua/jevmeter</a></p>Scoring & Rankinghttps://jevbooks.com/en/p/sorrycc__typesafe-snakehttps://jevbooks.com/en/p/sorrycc__typesafe-snakeJev picks one legal Snake move per tick.Sun, 20 Sep 2026 08:43:12 GMT<p>Jev picks one legal Snake move per tick.</p><p>A TypeScript web app where TypeSafe's Jev, a System One decision model, auto-plays Snake in real time. Built for developers exploring typed, non-LLM decisions in a game loop.</p><p><a href="https://github.com/sorrycc/typesafe-snake">sorrycc/typesafe-snake</a></p>Game & Simulationfailure-mode-knobhttps://jevbooks.com/en/p/monteduro__killmyideahttps://jevbooks.com/en/p/monteduro__killmyideaDescribe a startup idea; Jev decides KILL, FIX, or SHIP.Sun, 20 Sep 2026 08:43:06 GMT<p>Describe a startup idea; Jev decides KILL, FIX, or SHIP.</p><p>A web app that evaluates startup ideas for indie hackers. You describe an idea, and it returns a verdict — KILL IT, FIX IT, or SHIP IT — plus scores, a best signal and biggest risk.</p><p><a href="https://github.com/monteduro/killmyidea">monteduro/killmyidea</a></p>Evaluation & Benchmarkingflat-fanouttiered-thresholdscode-owned-policyhttps://jevbooks.com/en/p/skyvern-ai__jevscapehttps://jevbooks.com/en/p/skyvern-ai__jevscapeRuneBench harness for TypeSafe's Jev to play RuneScape.Sun, 20 Sep 2026 08:42:58 GMT<p>RuneBench harness for TypeSafe's Jev to play RuneScape.</p><p>A RuneBench fork that lets TypeSafe's Jev, a non-generative model, play RuneScape through a bounded action catalog built on rs-sdk. It records tick-level results and provides a live dashboard for developers and benchmarkers.</p><p><a href="https://github.com/Skyvern-AI/jevscape">Skyvern-AI/jevscape</a></p>Game & Simulationflat-fanouthttps://jevbooks.com/en/p/standardagents__jevpilothttps://jevbooks.com/en/p/standardagents__jevpilotThree.js driving simulator with Jev-powered autopilotSun, 20 Sep 2026 08:42:51 GMT<p>Three.js driving simulator with Jev-powered autopilot</p><p>A playable Three.js driving simulator where Jev chooses among candidate paths and speeds while local code handles vehicle dynamics and geometry. For developers exploring AI-driven driving decisions.</p><p><a href="https://github.com/standardagents/jevpilot">standardagents/jevpilot</a></p>Game & Simulationhttps://jevbooks.com/en/p/thelau__jev-tetrishttps://jevbooks.com/en/p/thelau__jev-tetrisTetris played by a judgment model that cannot write or see.Sun, 20 Sep 2026 08:42:48 GMT<p>Tetris played by a judgment model that cannot write or see.</p><p>A Tetris game where deterministic code enumerates every reachable placement, including tucks and spins, and writes each as an English sentence. Jev returns a probability for all options across five Choice questions; the highest-weighted option is played and the whole distribution is drawn on the board.</p><p><a href="https://github.com/thelau/jev-tetris">thelau/jev-tetris</a></p>Game & Simulationcode-owned-policyhttps://jevbooks.com/en/p/amycardoso__jev-palettehttps://jevbooks.com/en/p/amycardoso__jev-paletteSemantic command palette ranked by intent with TypeSafe Jev.Sun, 20 Sep 2026 08:42:46 GMT<p>Semantic command palette ranked by intent with TypeSafe Jev.</p><p>A demo command palette (⌘K) for a fictional mail app. It ranks 77 commands by meaning from Portuguese or English intent, side by side with a classic fuzzy matcher that fails on the same queries.</p><p><a href="https://github.com/amycardoso/jev-palette">amycardoso/jev-palette</a></p>Scoring & Rankingfailure-mode-knobmeasured-calibrationhttps://jevbooks.com/en/p/iamvatsalpatel__tiershifthttps://jevbooks.com/en/p/iamvatsalpatel__tiershiftPolicy-bounded model routing for TypeScript and Python.Sun, 20 Sep 2026 08:42:23 GMT<p>Policy-bounded model routing for TypeScript and Python.</p><p>tiershift reads each LLM request, picks the cheapest model tier that can handle it, and escalates on evidence. It is for developers who want to cut API costs without losing quality.</p><p><a href="https://github.com/iamvatsalpatel/tiershift">iamvatsalpatel/tiershift</a></p>Classification & Routingscreen-then-deephttps://jevbooks.com/en/p/kyotofin__tax-doc-classifierhttps://jevbooks.com/en/p/kyotofin__tax-doc-classifierJev-based classifier for IRS tax document pages.Sun, 20 Sep 2026 08:42:19 GMT<p>Jev-based classifier for IRS tax document pages.</p><p>Classifies text-bearing PDF pages into IRS form and page-kind candidates with a confidence gate, for teams triaging tax documents. It is document triage, not tax advice, and scanned pages need OCR first.</p><p><a href="https://github.com/kyotofin/tax-doc-classifier">kyotofin/tax-doc-classifier</a></p>Classification & Routingmeasured-calibrationhttps://jevbooks.com/en/p/giuliosmall__pg_typesafehttps://jevbooks.com/en/p/giuliosmall__pg_typesafePostgreSQL extension calling TypeSafe AI (Jev) from SQL for classification.Sun, 20 Sep 2026 08:42:18 GMT<p>PostgreSQL extension calling TypeSafe AI (Jev) from SQL for classification.</p><p>A pre-alpha PostgreSQL C extension that exposes TypeSafe AI (System One / Jev) categorical judgments from SQL. It is for PostgreSQL users who want Choice, Noul, and Score decisions on text without leaving the database.</p><p><a href="https://github.com/giuliosmall/pg_typesafe">giuliosmall/pg_typesafe</a></p>Classification & Routingflat-fanouthttps://jevbooks.com/en/p/realzachi__pg-jevhttps://jevbooks.com/en/p/realzachi__pg-jevAsk Postgres tables plain-language questions with calibrated probabilities.Sun, 20 Sep 2026 08:42:10 GMT<p>Ask Postgres tables plain-language questions with calibrated probabilities.</p><p>A PostgreSQL extension that lets you filter, rank and classify table rows with plain-language conditions, using TypeSafe's Jev model instead of embeddings or vector columns. Aimed at SQL users who want semantic queries over ordinary tables.</p><p><a href="https://github.com/realZachi/pg-jev">realZachi/pg-jev</a></p>Infra / SDKs / Integrationsflat-fanouthttps://jevbooks.com/en/p/n3ndor__n8n-nodes-typesafe-jevhttps://jevbooks.com/en/p/n3ndor__n8n-nodes-typesafe-jevn8n community node for TypeSafe Jev structured AI decisionsSun, 20 Sep 2026 08:42:09 GMT<p>n8n community node for TypeSafe Jev structured AI decisions</p><p>An n8n community node that lets workflows ask several typed questions about the same state and receive calibrated probabilities for classification, routing, verification and scoring. For n8n users who need structured decisions instead of generated prose.</p><p><a href="https://github.com/n3ndor/n8n-nodes-typesafe-jev">n3ndor/n8n-nodes-typesafe-jev</a></p>Classification & Routingflat-fanouthttps://jevbooks.com/en/p/eugeneboondock__jevsqlhttps://jevbooks.com/en/p/eugeneboondock__jevsqlSQL with natural-language predicates, powered by TypeSafe's Jev.Sun, 20 Sep 2026 08:42:07 GMT<p>SQL with natural-language predicates, powered by TypeSafe's Jev.</p><p>JevSQL adds TypeSafe Jev judgments to SQLite, letting you compare records by meaning, select exact evidence from text, rank retrieved passages, route uncertain results to review, and detect when a changed source invalidates an earlier decision. It runs as a Node library or CLI with zero runtime dependencies.</p><p><a href="https://github.com/EugeneBoondock/jevsql">EugeneBoondock/jevsql</a></p>Scoring & Rankinghttps://jevbooks.com/en/p/reachjalil__jevlogshttps://jevbooks.com/en/p/reachjalil__jevlogsScore OpenTelemetry logs before expensive LLM analysis.Sun, 20 Sep 2026 08:42:04 GMT<p>Score OpenTelemetry logs before expensive LLM analysis.</p><p>Jev Logs is a local OpenTelemetry log receiver that scores each log's diagnostic value, priority, and routing decision so that costly reasoning-model analysis is only spent on records worth investigating.</p><p><a href="https://github.com/reachjalil/jevlogs">reachjalil/jevlogs</a></p>Classification & Routingscreen-then-deephttps://jevbooks.com/en/p/reachjalil__jev-treehttps://jevbooks.com/en/p/reachjalil__jev-treeRecursive Jev choice over taxonomies larger than 255 options.Sun, 20 Sep 2026 08:42:02 GMT<p>Recursive Jev choice over taxonomies larger than 255 options.</p><p>jev-tree lets you select among thousands of leaves by walking a JSON tree and calling Jev once per level or partition, working around TypeSafe Jev's 255-option choice cap. For developers with large catalogs.</p><p><a href="https://github.com/reachjalil/jev-tree">reachjalil/jev-tree</a></p>Infra / SDKs / Integrationsfailure-mode-knobhttps://jevbooks.com/en/p/jarrodwatts__jev-traderhttps://jevbooks.com/en/p/jarrodwatts__jev-traderOne AI trade decision every Monad block on Kuru MON-USDC.Sun, 20 Sep 2026 08:42:00 GMT<p>One AI trade decision every Monad block on Kuru MON-USDC.</p><p>A TypeScript trading bot that posts a post-only limit order on the Kuru MON-USDC order book every Monad block, using a TypeSafe Jev model to decide buy or sell. It dry-runs without a private key.</p><p><a href="https://github.com/jarrodwatts/jev-trader">jarrodwatts/jev-trader</a></p>Finance & Tradinghttps://jevbooks.com/en/p/choxos__jev-reviewerhttps://jevbooks.com/en/p/choxos__jev-reviewerVerbatim quote extraction for systematic reviews, verified by Jev.Sun, 20 Sep 2026 08:41:57 GMT<p>Verbatim quote extraction for systematic reviews, verified by Jev.</p><p>A browser-based extraction aid for systematic reviewers. Open a trial report and its supplements, ask your extraction form or a RoB 2, ROBINS-I, QUADAS-2 or TIDieR template, and get verbatim quotes with file and page for human checking.</p><p><a href="https://github.com/choxos/jev-reviewer">choxos/jev-reviewer</a></p>Data Labeling & Curationcross-checkcandidate-enumerationhttps://jevbooks.com/en/p/colliber__duckdb-jevhttps://jevbooks.com/en/p/colliber__duckdb-jevDuckDB extension: typed Jev answers as real SQL typesSun, 20 Sep 2026 08:41:53 GMT<p>DuckDB extension: typed Jev answers as real SQL types</p><p>A DuckDB extension that lets you ask a Jev question about every row of a table or Parquet file in SQL, returning real SQL types instead of text.</p><p><a href="https://github.com/colliber/duckdb-jev">colliber/duckdb-jev</a></p>Infra / SDKs / Integrationsflat-fanouthttps://jevbooks.com/en/p/awlevin__typesafe-computer-usehttps://jevbooks.com/en/p/awlevin__typesafe-computer-usemacOS computer use via OCR and bounded Jev action selection.Sun, 20 Sep 2026 08:41:52 GMT<p>macOS computer use via OCR and bounded Jev action selection.</p><p>A macOS computer-use loop that drives a Mac toward a plain-English goal. It reads the screen deterministically and asks a small classifier for the next action, avoiding screenshot calls to large models.</p><p><a href="https://github.com/awlevin/typesafe-computer-use">awlevin/typesafe-computer-use</a></p>Agent Decisionshttps://jevbooks.com/en/p/droidrun__mobile-jevhttps://jevbooks.com/en/p/droidrun__mobile-jevAndroid agent where Jev picks every action from live device state.Sun, 20 Sep 2026 08:41:52 GMT<p>Android agent where Jev picks every action from live device state.</p><p>A standalone mobile agent for Mobilerun that controls a real Android device from a goal, with a React studio, CLI, traces, and latency measurements. Built for developers exploring bounded, typed model decisions on phones.</p><p><a href="https://github.com/droidrun/mobile-jev">droidrun/mobile-jev</a></p>Agent Decisionshttps://jevbooks.com/en/p/moritzkremb__jev-voice-browserhttps://jevbooks.com/en/p/moritzkremb__jev-voice-browserVoice control for a real browser, deciding intent and target with Jev.Sun, 20 Sep 2026 08:41:43 GMT<p>Voice control for a real browser, deciding intent and target with Jev.</p><p>A Node app that drives a headed Chromium window via Playwright using spoken commands. It streams partial Web Speech API transcripts to a local server, which asks Jev for typed probabilities and acts, waits, asks, or ignores.</p><p><a href="https://github.com/moritzkremb/jev-voice-browser">moritzkremb/jev-voice-browser</a></p>Agent Decisionsflat-fanoutcandidate-enumerationhttps://jevbooks.com/en/p/wy-coliney__jev-browser-usehttps://jevbooks.com/en/p/wy-coliney__jev-browser-useJev drives browser controls; Codex types and verifies.Sun, 20 Sep 2026 08:41:42 GMT<p>Jev drives browser controls; Codex types and verifies.</p><p>A Codex browser Skill for developers running repetitive browser workflows: dashboards, settings, reports and checks. Jev handles navigation, clicks, toggles and scrolling; Codex handles text input, visual judgment and the final verification. Built while testing EZCollegeApp.</p><p><a href="https://github.com/wy-coliney/jev-browser-use">wy-coliney/jev-browser-use</a></p>Agent Decisionshttps://jevbooks.com/en/p/ghalebdweikat__winnowhttps://jevbooks.com/en/p/ghalebdweikat__winnowCalibrated context sieve that judges tool results before they enter Claude Code context.Sun, 20 Sep 2026 08:41:39 GMT<p>Calibrated context sieve that judges tool results before they enter Claude Code context.</p><p>A Claude Code plugin for developers running long agent sessions. It intercepts large Read, Bash, and Grep results, splits them into ~25-line blocks, and replaces blocks the judge is confident you don't need with a three-line stub that can be restored on demand.</p><p><a href="https://github.com/GhalebDweikat/winnow">GhalebDweikat/winnow</a></p>Verification & Guardrailsflat-fanouthttps://jevbooks.com/en/p/kevinpita__pi-jev-contexthttps://jevbooks.com/en/p/kevinpita__pi-jev-contextReversible context pruning for Pi, powered by TypeSafe Jev.Sun, 20 Sep 2026 08:41:38 GMT<p>Reversible context pruning for Pi, powered by TypeSafe Jev.</p><p>An opt-in Pi extension that filters older messages from model requests while preserving the original session history. It is for Pi users running long sessions who want to keep useful context without deleting history.</p><p><a href="https://github.com/kevinpita/pi-jev-context">kevinpita/pi-jev-context</a></p>Scoring & Rankinghttps://jevbooks.com/en/p/ilkerulusoy__pi-jev-compacthttps://jevbooks.com/en/p/ilkerulusoy__pi-jev-compactSelective, verbatim context compaction for Pi using Jev.Sun, 20 Sep 2026 08:41:33 GMT<p>Selective, verbatim context compaction for Pi using Jev.</p><p>A Pi extension for developers running long sessions. Instead of asking an LLM to summarize old turns, it asks Jev to score each tool call and drops only what Jev says is no longer needed.</p><p><a href="https://github.com/ilkerulusoy/pi-jev-compact">ilkerulusoy/pi-jev-compact</a></p>Agent Decisionsflat-fanouttiered-thresholdshttps://jevbooks.com/en/p/dicklesworthstone__skillrankerhttps://jevbooks.com/en/p/dicklesworthstone__skillrankerRust CLI ranking agent skills for the next step with Jev.Sun, 20 Sep 2026 08:41:30 GMT<p>Rust CLI ranking agent skills for the next step with Jev.</p><p>A standalone Rust CLI for developers running agents with large skill libraries. It ranks which skills fit the next step using live session context, and can abstain when none fit.</p><p><a href="https://github.com/Dicklesworthstone/skillranker">Dicklesworthstone/skillranker</a></p>Agent Decisionshttps://jevbooks.com/en/p/lakeday-org__perchhttps://jevbooks.com/en/p/lakeday-org__perchSemantic code linting with Jev.Sun, 20 Sep 2026 08:41:26 GMT<p>Semantic code linting with Jev.</p><p>Perch is a semantic code linter for developers. It scans code units, evaluates them against configurable Jev questions, and reports problems with severity, type, confidence, and method.</p><p><a href="https://github.com/lakeday-org/perch">lakeday-org/perch</a></p>Evaluation & Benchmarkinghttps://jevbooks.com/en/p/aaronshaf__opencode-jev-orchestratorhttps://jevbooks.com/en/p/aaronshaf__opencode-jev-orchestratorSticky cheap OpenCode parent; Jev escalates hard turns to subagents.Sun, 20 Sep 2026 08:41:19 GMT<p>Sticky cheap OpenCode parent; Jev escalates hard turns to subagents.</p><p>An OpenCode orchestrator plugin for OpenCode Go users that keeps the parent session on a cheap sticky model for warm cache and delegates harder turns to temporary stronger child subagents.</p><p><a href="https://github.com/aaronshaf/opencode-jev-orchestrator">aaronshaf/opencode-jev-orchestrator</a></p>Classification & Routinghysteresishttps://jevbooks.com/en/p/brainwires__jevwirehttps://jevbooks.com/en/p/brainwires__jevwireJev decision layer for agents: MCP tools, library, Claude Code hooks.Sun, 20 Sep 2026 08:41:16 GMT<p>Jev decision layer for agents: MCP tools, library, Claude Code hooks.</p><p>jevwire wires TypeSafe AI's Jev classifier into agent harnesses. It offers 6 MCP tools, an embeddable JevDecisionModel library, and a Claude Code plugin whose hooks judge tool calls, fetched results, and turn endings.</p><p><a href="https://github.com/Brainwires/jevwire">Brainwires/jevwire</a></p>Agent Decisionsflat-fanouthttps://jevbooks.com/en/p/kierandotai__jev-scouthttps://jevbooks.com/en/p/kierandotai__jev-scoutJev-scored observable web research for MCP agentsSun, 20 Sep 2026 08:41:13 GMT<p>Jev-scored observable web research for MCP agents</p><p>An MCP stdio server exposing guarded_search, guarded_fetch, and research_report for agents that research the web. It annotates or gates search results and fetched pages, and streams every decision to a local dashboard.</p><p><a href="https://github.com/kierandotai/jev-scout">kierandotai/jev-scout</a></p>Scoring & Rankinghttps://jevbooks.com/en/p/billionsbobby__jevrouterhttps://jevbooks.com/en/p/billionsbobby__jevrouterJev-powered router for models, tools, and subagents.Sun, 20 Sep 2026 08:41:09 GMT<p>Jev-powered router for models, tools, and subagents.</p><p>JevRouter routes agent requests across models, subagents, skills, MCP tools, CLIs, and plugins. It uses Jev Choice decisions to pick the next capability, while the host enforces availability, permissions, risk, and confirmation before execution.</p><p><a href="https://github.com/BillionsBobby/JevRouter">BillionsBobby/JevRouter</a></p>Classification & Routingmeasured-calibrationhttps://jevbooks.com/en/p/niazmorshed2007__jev-reviewhttps://jevbooks.com/en/p/niazmorshed2007__jev-reviewLocal MCP server giving coding agents Jev quality scores.Sun, 20 Sep 2026 08:41:04 GMT<p>Local MCP server giving coding agents Jev quality scores.</p><p>Jev Review is a local-first MCP plugin that gives Claude Code, Codex, Cursor, and OpenCode structured software-quality scores while they work, across correctness, complexity, changeability, modularity, tests, security, and other dimensions.</p><p><a href="https://github.com/NiazMorshed2007/jev-review">NiazMorshed2007/jev-review</a></p>Scoring & Rankinghttps://jevbooks.com/en/p/0xnatoshi__jev-codex-routerhttps://jevbooks.com/en/p/0xnatoshi__jev-codex-routerPer-turn Codex model, reasoning, and speed-mode routing via Jev.Sun, 20 Sep 2026 08:41:01 GMT<p>Per-turn Codex model, reasoning, and speed-mode routing via Jev.</p><p>A router extension for an existing local Codex Router installation. It picks a model and thinking effort for each Codex model call, including continuations after tools, aiming for sufficient capability without unnecessary quota use.</p><p><a href="https://github.com/0xNatoshi/jev-codex-router">0xNatoshi/jev-codex-router</a></p>Classification & Routingfailure-mode-knobhttps://jevbooks.com/en/p/devanshbatham__commit-minerhttps://jevbooks.com/en/p/devanshbatham__commit-minerClassify Git commits with Jev: change types, bug and security fixes.Sun, 20 Sep 2026 08:40:58 GMT<p>Classify Git commits with Jev: change types, bug and security fixes.</p><p>commit-miner is a Rust CLI that scans Git repositories and classifies commit diffs and messages into change types and candidate security-fix/CWE labels for inspection. It is aimed at developers and security reviewers auditing commit history.</p><p><a href="https://github.com/devanshbatham/commit-miner">devanshbatham/commit-miner</a></p>Classification & Routinghttps://jevbooks.com/en/p/qkal__cannyhttps://jevbooks.com/en/p/qkal__cannyEvidence ledger that blocks unsupported "done" claims from AI coding agents.Sun, 20 Sep 2026 08:40:58 GMT<p>Evidence ledger that blocks unsupported &quot;done&quot; claims from AI coding agents.</p><p>Canny is a warden for AI coding agents. It hooks into Claude Code and Codex CLI, keeps an append-only ledger of what the agent actually did, and refuses to let the agent finish on an unsupported claim.</p><p><a href="https://github.com/qkal/Canny">qkal/Canny</a></p>Verification & Guardrailshttps://jevbooks.com/en/p/ellipsis-dev__blinkhttps://jevbooks.com/en/p/ellipsis-dev__blinkJev-guided walkers search a codebase for a natural-language query.Sun, 20 Sep 2026 08:40:53 GMT<p>Jev-guided walkers search a codebase for a natural-language query.</p><p>Blink is a TypeScript CLI that finds files in a codebase matching a natural-language query. It is aimed at developers who want to locate code by description rather than by exact path or filename.</p><p><a href="https://github.com/ellipsis-dev/blink">ellipsis-dev/blink</a></p>Agent Decisionshttps://jevbooks.com/en/p/nidhi-singh02__agent-routerhttps://jevbooks.com/en/p/nidhi-singh02__agent-routerQuota-aware CLI routing of coding tasks to AI agents.Sun, 20 Sep 2026 08:40:50 GMT<p>Quota-aware CLI routing of coding tasks to AI agents.</p><p>A local-first CLI that picks an AI coding agent, model, and reasoning effort for a coding phase, then launches it in a Herdr pane. Built for developers managing multiple agent subscriptions and shared accounts.</p><p><a href="https://github.com/nidhi-singh02/agent-router">nidhi-singh02/agent-router</a></p>Scoring & Rankinghttps://jevbooks.com/en/p/jomatsu__zod-jevhttps://jevbooks.com/en/p/jomatsu__zod-jevZod validates shape; Jev validates meaning.Sun, 20 Sep 2026 08:40:50 GMT<p>Zod validates shape; Jev validates meaning.</p><p>zod-jev composes TypeSafe Jev semantic checks into Zod 4 schemas, so TypeScript projects can turn model-only judgments about free text into calibrated probabilities their code can threshold on.</p><p><a href="https://github.com/jomatsu/zod-jev">jomatsu/zod-jev</a></p>Verification & Guardrailsflat-fanoutfailure-mode-knobhttps://jevbooks.com/en/p/zhirschtritt__typesafe-gohttps://jevbooks.com/en/p/zhirschtritt__typesafe-goUnofficial Go SDK for the TypeSafe v1 API.Sun, 20 Sep 2026 08:40:46 GMT<p>Unofficial Go SDK for the TypeSafe v1 API.</p><p>A small, dependency-free Go client for the TypeSafe v1 API, covering System One calls and model discovery. Aimed at Go developers who want to call TypeSafe from Go without pulling in extra dependencies.</p><p><a href="https://github.com/zhirschtritt/typesafe-go">zhirschtritt/typesafe-go</a></p>Infra / SDKs / Integrationsflat-fanouthttps://jevbooks.com/en/p/sharziki__semdecidehttps://jevbooks.com/en/p/sharziki__semdecideTyped semantic decisions for Unix pipelines and CI, powered by Jev.Sun, 20 Sep 2026 08:40:45 GMT<p>Typed semantic decisions for Unix pipelines and CI, powered by Jev.</p><p>SemDecide is a CLI for Unix pipelines and CI that turns text or JSONL into semantic predicates, routes, scores, and filtered streams, with stable exit codes, for shell scripts and automation.</p><p><a href="https://github.com/sharziki/semdecide">sharziki/semdecide</a></p>Classification & Routinghttps://jevbooks.com/en/p/simxnherrera__jevrhttps://jevbooks.com/en/p/simxnherrera__jevrNative R client for Jev System One decisionsSun, 20 Sep 2026 08:40:42 GMT<p>Native R client for Jev System One decisions</p><p>jevr is an R client for JEV / TypeSafe System One. It sends a state and typed questions to a model and returns structured answers for use in R, supporting multiple questions, probabilities, confidence, usage, and provider metadata.</p><p><a href="https://github.com/simxnherrera/jevr">simxnherrera/jevr</a></p>Infra / SDKs / Integrationsflat-fanouthttps://jevbooks.com/en/p/abovecolin__jevclienthttps://jevbooks.com/en/p/abovecolin__jevclientAsync Python client for TypeSafe Jev typed questions.Sun, 20 Sep 2026 08:40:38 GMT<p>Async Python client for TypeSafe Jev typed questions.</p><p>An independent async Python client for TypeSafe Jev, the System One decision model. It lets Python developers ask typed questions about a state and receive probabilities and choices instead of prose.</p><p><a href="https://github.com/AboveColin/jevclient">AboveColin/jevclient</a></p>Infra / SDKs / Integrationsflat-fanouthttps://jevbooks.com/en/p/mrnugget__jev-shell-historyhttps://jevbooks.com/en/p/mrnugget__jev-shell-historyFish-style zsh history autosuggestions ranked by Jev.Sun, 20 Sep 2026 08:40:34 GMT<p>Fish-style zsh history autosuggestions ranked by Jev.</p><p>A zsh plugin that shows Fish-style inline autosuggestions. As you type, it ranks your last 100 distinct history entries and displays the best match in grey after the cursor, for zsh 5.9+ users with a TypeSafe API key.</p><p><a href="https://github.com/mrnugget/jev-shell-history">mrnugget/jev-shell-history</a></p>Scoring & Rankingtiered-thresholdshttps://jevbooks.com/en/p/byk__jev-mcphttps://jevbooks.com/en/p/byk__jev-mcpEval-first MCP server for Jev's typed judgments with probabilities.Sun, 20 Sep 2026 08:40:30 GMT<p>Eval-first MCP server for Jev's typed judgments with probabilities.</p><p>An MCP server for TypeSafe's Jev, a System One model that returns typed judgments (noul, choice, score) with probabilities instead of generated text. It helps developers prototype, evaluate, and apply questions to many items.</p><p><a href="https://github.com/BYK/jev-mcp">BYK/jev-mcp</a></p>Evaluation & Benchmarkingflat-fanoutmeasured-calibrationhttps://jevbooks.com/en/p/inanna-malick__jev-dslhttps://jevbooks.com/en/p/inanna-malick__jev-dslTyped Haskell DSL for TypeSafe Jev judgment packets.Sun, 20 Sep 2026 08:40:25 GMT<p>Typed Haskell DSL for TypeSafe Jev judgment packets.</p><p>An early-alpha Haskell DSL for agents: a packet of labelled questions is written once as an expression, its type is inferred, it renders to Jev request JSON, and answers return under the same labels as records read by field.</p><p><a href="https://github.com/inanna-malick/jev-dsl">inanna-malick/jev-dsl</a></p>Agent Decisionsflat-fanouttiered-thresholdshttps://jevbooks.com/en/p/rupeshpoojary9__poorjevhttps://jevbooks.com/en/p/rupeshpoojary9__poorjevLocal, open-source Jev alternative with calibrated confidence.Sun, 20 Sep 2026 08:40:00 GMT<p>Local, open-source Jev alternative with calibrated confidence.</p><p>poorjev is a local-first System One decision layer for LLM apps. It provides typed decisions (Choice, Score, Noul) with provably calibrated confidence, runs offline on commodity zero-shot NLI models, and requires no API key or waitlist.</p><p><a href="https://github.com/rupeshpoojary9/poorjev">rupeshpoojary9/poorjev</a></p>Classification & Routingflat-fanouthttps://jevbooks.com/en/p/theoleecj__semifhttps://jevbooks.com/en/p/theoleecj__semifSemantic ifs from open models, on a 3090 at home.Sun, 20 Sep 2026 08:39:57 GMT<p>Semantic ifs from open models, on a 3090 at home.</p><p>SemIf is an independent open-model research baseline for direct typed option scoring. It reproduces the interface pattern of runtime-defined semantic decisions, not Jev's undisclosed model or training. For developers running semantic decisions on a local GPU.</p><p><a href="https://github.com/TheoLeeCJ/SemIf">TheoLeeCJ/SemIf</a></p>Agent Decisionshttps://jevbooks.com/en/p/scienthoon__lucehttps://jevbooks.com/en/p/scienthoon__luceOpen recipe for Jev-style decision models: task description to calibrated probabilities.Sun, 20 Sep 2026 08:39:56 GMT<p>Open recipe for Jev-style decision models: task description to calibrated probabilities.</p><p>Luce is an open recipe for building Jev-style decision models. It scaffolds questions, generates or annotates data with an LLM teacher, trains a LoRA + decision head on Qwen, evaluates calibration, and serves probabilities.</p><p><a href="https://github.com/scienthoon/luce">scienthoon/luce</a></p>Calibration & Researchrules-as-questionsmeasured-calibrationhttps://jevbooks.com/en/p/anessbelbati__jev-rerank-benchhttps://jevbooks.com/en/p/anessbelbati__jev-rerank-benchBenchmarking Jev vs dedicated rerankers on 14 datasets.Sun, 20 Sep 2026 08:39:54 GMT<p>Benchmarking Jev vs dedicated rerankers on 14 datasets.</p><p>A benchmark comparing TypeSafe Jev against Cohere Rerank 4, ZeroEntropy zerank-2, and a chat-model baseline for passage reranking, with raw API responses, scoring code, and uncertainty intervals.</p><p><a href="https://github.com/anessbelbati/jev-rerank-bench">anessbelbati/jev-rerank-bench</a></p>Scoring & Rankingstate-minimalismhttps://jevbooks.com/en/p/firassx914__janushttps://jevbooks.com/en/p/firassx914__janusMeasure where to route between Jev and a frontier model.Sun, 20 Sep 2026 08:39:52 GMT<p>Measure where to route between Jev and a frontier model.</p><p>Janus measures, on your own labelled data or decision logs, whether routing between a small model and a larger one pays off, and where the threshold sits. It ships no default threshold.</p><p><a href="https://github.com/FirasSX914/Janus">FirasSX914/Janus</a></p>Classification & Routingscreen-then-deephttps://jevbooks.com/en/p/anxkhn__jevplayspokemonhttps://jevbooks.com/en/p/anxkhn__jevplayspokemonLevel-40 Pokémon agent using Jev for battle decisions on real FireRed.Sun, 20 Sep 2026 08:39:49 GMT<p>Level-40 Pokémon agent using Jev for battle decisions on real FireRed.</p><p>A Pokémon-playing agent that runs Gen 3 battles in two modes: Pokémon Showdown text battles on any computer, and a real FireRed ROM on Mac with Apple Silicon. Jev picks each turn's move or switch.</p><p><a href="https://github.com/anxkhn/JevPlaysPokemon">anxkhn/JevPlaysPokemon</a></p>Game & Simulationhttps://jevbooks.com/en/p/abdelstark__heist-onehttps://jevbooks.com/en/p/abdelstark__heist-oneObservable browser stealth game where Jev judges, deterministic code owns the world.Sun, 20 Sep 2026 08:39:47 GMT<p>Observable browser stealth game where Jev judges, deterministic code owns the world.</p><p>HEIST//ONE is a browser stealth game for players and developers: six guards, three access-controlled doors, disguises, credentials, light control, noise decoys, one artifact, and two exits, with an authoritative server simulation and a Decision Lens.</p><p><a href="https://github.com/AbdelStark/heist-one">AbdelStark/heist-one</a></p>Game & Simulationflat-fanouthttps://jevbooks.com/en/p/openbyteinc__quantdingerhttps://jevbooks.com/en/p/openbyteinc__quantdingerOpen-source AI trading OS with Jev System One integration.Sun, 20 Sep 2026 08:39:45 GMT<p>Open-source AI trading OS with Jev System One integration.</p><p>QuantDinger is a self-hosted quantitative trading platform for independent traders, Python strategy authors, and small teams. It supports AI research, Python strategy development, backtesting, paper/live trading across crypto, stocks, and forex, and multi-tenant SaaS deployment.</p><p><a href="https://github.com/OpenByteInc/QuantDinger">OpenByteInc/QuantDinger</a></p>Finance & Tradinghttps://jevbooks.com/en/p/itsmostafa__typesafe-mcphttps://jevbooks.com/en/p/itsmostafa__typesafe-mcpMCP server giving AI agents typed Jev decisions over stdio.Sun, 20 Sep 2026 08:39:30 GMT<p>MCP server giving AI agents typed Jev decisions over stdio.</p><p>A Go CLI and single-binary MCP server, evaluate, that lets Claude Code, Claude Desktop, Codex and pi call TypeSafe's Jev model and get back probabilities they can branch on.</p><p><a href="https://github.com/itsmostafa/typesafe-mcp">itsmostafa/typesafe-mcp</a></p>Agent Decisionshttps://jevbooks.com/en/p/devmortimer__pi-wardenhttps://jevbooks.com/en/p/devmortimer__pi-wardenPi guardrails that steer the agent instead of interrupting you.Sun, 20 Sep 2026 08:39:25 GMT<p>Pi guardrails that steer the agent instead of interrupting you.</p><p>pi-warden supervises Pi coding agents, catching risky actions, ignored rules, stuck loops, unverified done claims, security issues, runaway output, and wasted context. It usually feeds issues back to the agent rather than interrupting the user.</p><p><a href="https://github.com/DevMortimer/pi-warden">DevMortimer/pi-warden</a></p>Verification & Guardrailshttps://jevbooks.com/en/p/nyarlathoteppppp__pi-jev-contexthttps://jevbooks.com/en/p/nyarlathoteppppp__pi-jev-contextCache-neutral context trimming for pi, powered by TypeSafe Jev.Sun, 20 Sep 2026 08:39:24 GMT<p>Cache-neutral context trimming for pi, powered by TypeSafe Jev.</p><p>A pi coding agent extension that shortens long tool output before it enters the context, so cached prompt prefixes are never invalidated. For pi users who want smaller contexts without losing information.</p><p><a href="https://github.com/Nyarlathoteppppp/pi-jev-context">Nyarlathoteppppp/pi-jev-context</a></p>Agent Decisionsfailure-mode-knobhttps://jevbooks.com/en/p/kyu1204__jgrephttps://jevbooks.com/en/p/kyu1204__jgrepSemantic grep CLI that finds code by behavior using TypeSafe Jev.Sun, 20 Sep 2026 08:39:23 GMT<p>Semantic grep CLI that finds code by behavior using TypeSafe Jev.</p><p>jgrep is a semantic code search CLI for developers: it splits files or git diff hunks into 5–60 line chunks, asks Jev a yes/no question about each chunk, and prints file:line hits above a probability threshold, so diffs can be linted in CI with English rules.</p><p><a href="https://github.com/kyu1204/jgrep">kyu1204/jgrep</a></p>Scoring & Rankingflat-fanoutstate-minimalismhttps://jevbooks.com/en/p/antiyro__jevdroidhttps://jevbooks.com/en/p/antiyro__jevdroidControl Android over ADB with Jev-typed decisions.Sun, 20 Sep 2026 08:39:18 GMT<p>Control Android over ADB with Jev-typed decisions.</p><p>JevDroid is an experimental Python framework for controlling Android devices over ADB or UIAutomator2. Given a goal, it reads the accessibility tree, chooses an action, and executes it: open apps, tap, swipe, navigate back.</p><p><a href="https://github.com/antiyro/jevdroid">antiyro/jevdroid</a></p>Agent Decisionshttps://jevbooks.com/en/p/friedjof__jev-mobilehttps://jevbooks.com/en/p/friedjof__jev-mobileDurable Android agent using Jev for per-step action choices.Sun, 20 Sep 2026 08:39:18 GMT<p>Durable Android agent using Jev for per-step action choices.</p><p>Experimental autonomous Android sub-agent for a USB-connected device. Accepts a high-level task, persists it in SQLite, and runs observe → normalize → decide → mutate → verify. Currently a proof of concept, mainly tested against Android Settings and Keep/fixture scenarios.</p><p><a href="https://github.com/Friedjof/jev-mobile">Friedjof/jev-mobile</a></p>Agent Decisionshttps://jevbooks.com/en/p/nasrallah-al__jev-clihttps://jevbooks.com/en/p/nasrallah-al__jev-cliTyped AI judgments from the command line, powered by TypeSafe's Jev.Sun, 20 Sep 2026 08:39:18 GMT<p>Typed AI judgments from the command line, powered by TypeSafe's Jev.</p><p>A TypeScript CLI (installed as jevctl, command jev) that wraps TypeSafe's Jev model into pipeable, exit-code-gated shell commands for verifying claims, screening text, classifying, extracting, matching, routing, finding, reranking, compacting transcripts, and batching over JSONL.</p><p><a href="https://github.com/Nasrallah-AL/jev-cli">Nasrallah-AL/jev-cli</a></p>Verification & Guardrailshttps://jevbooks.com/en/p/sufianetaouil__everyhttps://jevbooks.com/en/p/sufianetaouil__everyAsk a yes/no question of every function; ranked probabilities in seconds.Sun, 20 Sep 2026 08:39:15 GMT<p>Ask a yes/no question of every function; ranked probabilities in seconds.</p><p>A semantic code search CLI for developers. It asks a yes/no question of every function in a codebase and ranks the resulting probabilities, so you can find code matching a described behavior rather than a text pattern.</p><p><a href="https://github.com/sufianetaouil/every">sufianetaouil/every</a></p>Scoring & Rankinghttps://jevbooks.com/en/p/numerous-com__dgphttps://jevbooks.com/en/p/numerous-com__dgpOpen protocol for decision-based AI agents with Jev-first orchestration.Sun, 20 Sep 2026 08:39:15 GMT<p>Open protocol for decision-based AI agents with Jev-first orchestration.</p><p>DGP is an open-source protocol by Numerous ApS for building decision-based AI agents. Applications expose evidence, typed decisions, and guarded actions; an agent assesses choices while application code retains control of permissions and effects.</p><p><a href="https://github.com/numerous-com/dgp">numerous-com/dgp</a></p>Agent Decisionshttps://jevbooks.com/en/p/abdelstark__bicameralhttps://jevbooks.com/en/p/abdelstark__bicameralHybrid coding harness: System 2 writes, System 1 (Jev) runs reflexes.Sun, 20 Sep 2026 08:39:11 GMT<p>Hybrid coding harness: System 2 writes, System 1 (Jev) runs reflexes.</p><p>A Pi coding harness where any LLM writes code while TypeSafe Jev supplies typed reflexes for policy, loop detection, and edit review. Aimed at developers who want calibrated S1 judgments without handing isolation to a model. Explicitly not a sandbox.</p><p><a href="https://github.com/AbdelStark/bicameral">AbdelStark/bicameral</a></p>Verification & Guardrailscode-owned-policyhttps://jevbooks.com/en/p/query-farm__vgi-typesafehttps://jevbooks.com/en/p/query-farm__vgi-typesafeTypeSafe System One questions as DuckDB LATERAL-joinable table functions.Sun, 20 Sep 2026 08:39:09 GMT<p>TypeSafe System One questions as DuckDB LATERAL-joinable table functions.</p><p>A VGI worker for DuckDB that exposes TypeSafe System One questions — choice, noul, and score — as table functions, so SQL users can classify, judge, and score rows in a LATERAL join or WHERE clause.</p><p><a href="https://github.com/Query-farm/vgi-typesafe">Query-farm/vgi-typesafe</a></p>Infra / SDKs / Integrationshttps://jevbooks.com/en/p/hawxy__typesafeai.nethttps://jevbooks.com/en/p/hawxy__typesafeai.net.NET SDK for TypeSafe AI's typed judgment API.Sun, 20 Sep 2026 08:39:08 GMT<p>.NET SDK for TypeSafe AI's typed judgment API.</p><p>A community .NET SDK for the TypeSafe AI System One API. It lets C# developers ask small, typed judgments about text or structured state and receive calibrated probabilities their code can act on.</p><p><a href="https://github.com/Hawxy/TypeSafeAI.Net">Hawxy/TypeSafeAI.Net</a></p>Infra / SDKs / Integrationsflat-fanouthttps://jevbooks.com/en/p/genierobot__typesafe-ai-railshttps://jevbooks.com/en/p/genierobot__typesafe-ai-railsCommunity Rails integration for TypeSafe's System One API.Sun, 20 Sep 2026 08:39:05 GMT<p>Community Rails integration for TypeSafe's System One API.</p><p>A community Rails gem that adds Rails configuration, persisted usage and cost telemetry, and opt-in confidence policies for TypeSafe's System One API, built on the typesafe-sdk Ruby gem.</p><p><a href="https://github.com/GenieRobot/typesafe-ai-rails">GenieRobot/typesafe-ai-rails</a></p>Infra / SDKs / Integrationsflat-fanouthttps://jevbooks.com/en/p/fox-islam__typesafe-sdk-phphttps://jevbooks.com/en/p/fox-islam__typesafe-sdk-phpPHP 8.3 SDK for TypeSafe's System One APISun, 20 Sep 2026 08:39:03 GMT<p>PHP 8.3 SDK for TypeSafe's System One API</p><p>An unofficial PHP client for TypeSafe AI's System One API, letting PHP and Laravel developers ask typed questions about text or JSON and receive probabilities, labels, or rubric levels.</p><p><a href="https://github.com/Fox-Islam/typesafe-sdk-php">Fox-Islam/typesafe-sdk-php</a></p>Classification & Routinghttps://jevbooks.com/en/p/ufec__typesafe-sdk-kotlinhttps://jevbooks.com/en/p/ufec__typesafe-sdk-kotlinCommunity Kotlin port of TypeSafe AI's JavaScript SDK for Android and JVM.Sun, 20 Sep 2026 08:39:02 GMT<p>Community Kotlin port of TypeSafe AI's JavaScript SDK for Android and JVM.</p><p>A Kotlin SDK for TypeSafe AI's System One API, targeting Android (minSdk 29) and the JVM with JDK 17. It is a community port of the JavaScript SDK 0.6.0, distributed via JitPack, and not affiliated with TypeSafe AI.</p><p><a href="https://github.com/ufec/typesafe-sdk-kotlin">ufec/typesafe-sdk-kotlin</a></p>Infra / SDKs / Integrationshttps://jevbooks.com/en/p/premo-cloud__typesafe-sdk-javahttps://jevbooks.com/en/p/premo-cloud__typesafe-sdk-javaCommunity Java 17 client for TypeSafe System One typed questions.Sun, 20 Sep 2026 08:39:02 GMT<p>Community Java 17 client for TypeSafe System One typed questions.</p><p>An unofficial community Java 17 client for the TypeSafe System One API. It lets Java developers ask typed Noul, Choice, and Score questions over application state and receive calibrated probabilities, mirroring the official Python and JavaScript SDKs.</p><p><a href="https://github.com/Premo-Cloud/typesafe-sdk-java">Premo-Cloud/typesafe-sdk-java</a></p>Infra / SDKs / Integrationsflat-fanouthttps://jevbooks.com/en/p/sergeax__typesafe-sdk-gohttps://jevbooks.com/en/p/sergeax__typesafe-sdk-goGo client for TypeSafe's System One API with typed questions.Sun, 20 Sep 2026 08:39:02 GMT<p>Go client for TypeSafe's System One API with typed questions.</p><p>A community Go 1.23 client for TypeSafe's System One API. It lets Go developers send a state plus typed Noul, Choice, and Score questions in one request and read back per-question answers.</p><p><a href="https://github.com/SergeAx/typesafe-sdk-go">SergeAx/typesafe-sdk-go</a></p>Infra / SDKs / Integrationsflat-fanouthttps://jevbooks.com/en/p/twister915__typesafe-aihttps://jevbooks.com/en/p/twister915__typesafe-aiTyped Rust client for TypeSafe's System One evaluation API.Sun, 20 Sep 2026 08:33:52 GMT<p>Typed Rust client for TypeSafe's System One evaluation API.</p><p>Brings TypeSafe's System One API into Rust as typed judgments for applications. It supports routing support cases, classifying content, scoring items on ordered rubrics, and judging conditions with probabilities. Provides async and blocking backends with observable retries and inspectable errors.</p><p><a href="https://github.com/Twister915/typesafe-ai">Twister915/typesafe-ai</a></p>Evaluation & Benchmarkingflat-fanouthttps://jevbooks.com/en/p/ticofab__scala-jev-sdkhttps://jevbooks.com/en/p/ticofab__scala-jev-sdkScala 3 client for Jev's typed questions, no effect system.Sun, 20 Sep 2026 08:33:47 GMT<p>Scala 3 client for Jev's typed questions, no effect system.</p><p>A community Scala 3 SDK for TypeSafe AI's Jev System One model, which answers typed questions about a state instead of generating text. It targets Scala developers who want routing, gating, ranking or guardrail decisions inside an application.</p><p><a href="https://github.com/ticofab/scala-jev-sdk">ticofab/scala-jev-sdk</a></p>Infra / SDKs / Integrationsflat-fanouthttps://jevbooks.com/en/p/abdelstark__s1-rshttps://jevbooks.com/en/p/abdelstark__s1-rsTyped System One decisions for Rust via derives and confidence gates.Sun, 20 Sep 2026 08:33:39 GMT<p>Typed System One decisions for Rust via derives and confidence gates.</p><p>s1-rs is a Rust workspace that turns fieldless enums and named structs into typed Choice, Score, and Noul questions for the TypeSafe System One API, returning compile-time-checked, confidence-gated answers. It targets Rust developers routing or classifying inputs.</p><p><a href="https://github.com/AbdelStark/s1-rs">AbdelStark/s1-rs</a></p>Classification & Routingtiered-thresholdshttps://jevbooks.com/en/p/kieranklaassen__ruby_llm-typesafehttps://jevbooks.com/en/p/kieranklaassen__ruby_llm-typesafeTypeSafe provider for RubyLLM 2 with typed, probabilistic answers.Sun, 20 Sep 2026 08:33:37 GMT<p>TypeSafe provider for RubyLLM 2 with typed, probabilistic answers.</p><p>A RubyLLM 2 provider for TypeSafe, which runs Jev, a System One model that returns probabilities for typed questions instead of text. For Ruby developers building structured-output applications.</p><p><a href="https://github.com/kieranklaassen/ruby_llm-typesafe">kieranklaassen/ruby_llm-typesafe</a></p>Infra / SDKs / Integrationsflat-fanouthttps://jevbooks.com/en/p/devmortimer__pi-typesafehttps://jevbooks.com/en/p/devmortimer__pi-typesafeTypeSafe decisions for Pi: batched evaluation, playground, typed API.Sun, 20 Sep 2026 08:33:34 GMT<p>TypeSafe decisions for Pi: batched evaluation, playground, typed API.</p><p>A Pi extension and library that gives the agent and other extensions one consented, key-managed TypeSafe client with a batched `typesafe_evaluate` tool and offline-testable transport; requests are billable and opt-in per user.</p><p><a href="https://github.com/DevMortimer/pi-typesafe">DevMortimer/pi-typesafe</a></p>Evaluation & Benchmarkingflat-fanouthttps://jevbooks.com/en/p/typesafe-ai__system-one-adapter-pythonhttps://jevbooks.com/en/p/typesafe-ai__system-one-adapter-pythonDrop-in TypeSafeClient replacement backed by LLM APIsSun, 20 Sep 2026 08:33:31 GMT<p>Drop-in TypeSafeClient replacement backed by LLM APIs</p><p>A Python adapter that replaces typesafe_sdk's system_one evaluation API with LLM APIs, letting developers compare TypeSafe against OpenAI, Anthropic, and OpenAI-compatible models on cost, speed, and intelligence.</p><p><a href="https://github.com/typesafe-ai/system-one-adapter-python">typesafe-ai/system-one-adapter-python</a></p>Evaluation & Benchmarkingflat-fanouthttps://jevbooks.com/en/p/marandaneto__typesafe-sdk-swifthttps://jevbooks.com/en/p/marandaneto__typesafe-sdk-swiftExperimental Swift SDK for the TypeSafe AI API.Sun, 20 Sep 2026 08:33:30 GMT<p>Experimental Swift SDK for the TypeSafe AI API.</p><p>An unofficial, experimental Swift-native client for the TypeSafe AI API, using Swift Package Manager, Swift 6 concurrency, and URLSession, for Apple-platform developers who want typed answers.</p><p><a href="https://github.com/marandaneto/typesafe-sdk-swift">marandaneto/typesafe-sdk-swift</a></p>Infra / SDKs / Integrationsflat-fanouthttps://jevbooks.com/en/p/jonesmelton__verdicthttps://jevbooks.com/en/p/jonesmelton__verdictTyped OCaml Eio client for TypeSafe AI's Jev evaluation APISun, 20 Sep 2026 08:33:21 GMT<p>Typed OCaml Eio client for TypeSafe AI's Jev evaluation API</p><p>An unofficial OCaml 5.2+ client SDK for the TypeSafe AI system-one evaluation API. It lets OCaml programs send a state plus typed questions and read back calibrated answers, with TLS, retries and bounded requests handled by the library.</p><p><a href="https://github.com/jonesmelton/verdict">jonesmelton/verdict</a></p>Evaluation & Benchmarkingflat-fanouthttps://jevbooks.com/en/p/typesafe-ai__typesafe-sdk-pythonhttps://jevbooks.com/en/p/typesafe-ai__typesafe-sdk-pythonOfficial Python client for the TypeSafe AI API.Sun, 20 Sep 2026 08:33:21 GMT<p>Official Python client for the TypeSafe AI API.</p><p>The official synchronous and asynchronous Python client for TypeSafe AI, letting Python developers call the TypeSafe API from their applications.</p><p><a href="https://github.com/typesafe-ai/typesafe-sdk-python">typesafe-ai/typesafe-sdk-python</a></p>Infra / SDKs / Integrationshttps://jevbooks.com/en/p/typesafe-ai__typesafe-sdk-jshttps://jevbooks.com/en/p/typesafe-ai__typesafe-sdk-jsOfficial TypeScript/JavaScript SDK for the TypeSafe AI API.Sun, 20 Sep 2026 08:33:15 GMT<p>Official TypeScript/JavaScript SDK for the TypeSafe AI API.</p><p>An official JavaScript and TypeScript client for the TypeSafe AI API, aimed at developers who want to call the System One model from Node.js 20 or newer with inferred answer types.</p><p><a href="https://github.com/typesafe-ai/typesafe-sdk-js">typesafe-ai/typesafe-sdk-js</a></p>Infra / SDKs / Integrationshttps://jevbooks.com/en/p/ethanplusai__jev-chat-for-twitchhttps://jevbooks.com/en/p/ethanplusai__jev-chat-for-twitchA Chrome extension that filters live Twitch chat with Jev.Sun, 20 Sep 2026 08:33:11 GMT<p>A Chrome extension that filters live Twitch chat with Jev.</p><p>A bring-your-own-key Chrome extension for Twitch viewers. It reads a channel's live chat over the anonymous IRC WebSocket and adds a second column showing only the messages matching a chosen intent: Helpful, Questions, Funny, Feedback, or Everything meaningful.</p><p><a href="https://github.com/ethanplusai/jev-chat-for-twitch">ethanplusai/jev-chat-for-twitch</a></p>Classification & Routinghttps://jevbooks.com/en/p/carldaws__hunchhttps://jevbooks.com/en/p/carldaws__hunchProbabilistic control flow for Ruby, powered by TypeSafe's Jev.Sun, 20 Sep 2026 08:32:45 GMT<p>Probabilistic control flow for Ruby, powered by TypeSafe's Jev.</p><p>Hunch lets Ruby developers ask plain-English questions about text or application data and branch on the answers. It provides chance, pick, and rate helpers, plus graded predicates, for use in Ruby and Rails apps.</p><p><a href="https://github.com/carldaws/hunch">carldaws/hunch</a></p>Classification & Routingflat-fanouttiered-thresholdshttps://jevbooks.com/en/p/prescott-data__jarviscore-frameworkhttps://jevbooks.com/en/p/prescott-data__jarviscore-frameworkPeer-to-peer multi-agent runtime with zero-trust credentials and durable state.Sun, 20 Sep 2026 08:32:44 GMT<p>Peer-to-peer multi-agent runtime with zero-trust credentials and durable state.</p><p>JarvisCore is a Python framework for building AI agent fleets that plan, reason, execute code, browse the web, and connect to external services. It targets developers who need crash recovery, shared memory, and peer-to-peer discovery without a central orchestrator.</p><p><a href="https://github.com/Prescott-Data/jarviscore-framework">Prescott-Data/jarviscore-framework</a></p>Agent Decisionshttps://jevbooks.com/en/p/zwlijay__jev-forgehttps://jevbooks.com/en/p/zwlijay__jev-forgeOpen stack for training and serving Jev-style decision models.Sun, 20 Sep 2026 08:32:39 GMT<p>Open stack for training and serving Jev-style decision models.</p><p>JevForge is an end-to-end toolkit for synthesizing decision data, training calibrated candidate scorers, evaluating them, and serving Jev-compatible inference for interactive web decisions. It is for researchers and developers building structured decision models.</p><p><a href="https://github.com/zwliJay/jev-forge">zwliJay/jev-forge</a></p>Agent Decisionsmeasured-calibrationhttps://jevbooks.com/en/p/replynodes__jev-web-analyzerhttps://jevbooks.com/en/p/replynodes__jev-web-analyzerSee what Jev thinks about your SaaS website.Sun, 20 Sep 2026 08:32:36 GMT<p>See what Jev thinks about your SaaS website.</p><p>A developer demo that analyzes a public SaaS landing page as clean Markdown and asks Jev ten bounded Choice questions about first-visit understanding, returning inspectable findings for the first change to make.</p><p><a href="https://github.com/replynodes/jev-web-analyzer">replynodes/jev-web-analyzer</a></p>Evaluation & Benchmarkingflat-fanouthttps://jevbooks.com/en/p/jyatesdotdev__jev-logtriagehttps://jevbooks.com/en/p/jyatesdotdev__jev-logtriageJev decides if a batch of logs is worth acting on.Sun, 20 Sep 2026 08:32:36 GMT<p>Jev decides if a batch of logs is worth acting on.</p><p>A Python tool for on-call operations that collapses Loki log batches into a single Jev call, then maps answers to suppress, watch, review, notify, or page. Nothing is executed.</p><p><a href="https://github.com/jyatesdotdev/jev-logtriage">jyatesdotdev/jev-logtriage</a></p>Classification & Routingflat-fanoutspeculative-headstiered-thresholdscode-owned-policyhttps://jevbooks.com/en/p/sutro-sh__jev-alignhttps://jevbooks.com/en/p/sutro-sh__jev-alignBuild calibrated AI Functions from human feedback using Jev and GEPA.Sun, 20 Sep 2026 08:32:36 GMT<p>Build calibrated AI Functions from human feedback using Jev and GEPA.</p><p>jev-align is an experimental CLI from Sutro for building AI Functions with TypeSafe's Jev. It finds uncertain examples, asks you to label them, and uses GEPA to improve the function.</p><p><a href="https://github.com/sutro-sh/jev-align">sutro-sh/jev-align</a></p>Data Labeling & Curationhttps://jevbooks.com/en/p/yuyang2230__jev-agent-skillhttps://jevbooks.com/en/p/yuyang2230__jev-agent-skillOffload agent classify, screen, score, and verify calls to Jev.Sun, 20 Sep 2026 08:32:36 GMT<p>Offload agent classify, screen, score, and verify calls to Jev.</p><p>A Claude Code / ZCode skill that moves high-frequency small judgments out of the main model's context and into Jev, cutting token cost for classification, batch screening, scoring, and compliance checks. Aimed at AI coding agents and their users.</p><p><a href="https://github.com/yuyang2230/jev-agent-skill">yuyang2230/jev-agent-skill</a></p>Classification & Routingflat-fanoutscreen-then-deepstate-minimalismhttps://jevbooks.com/en/p/milanboers__jev-plays-pokemonhttps://jevbooks.com/en/p/milanboers__jev-plays-pokemonJev picks Pokémon Red goals; code presses buttons.Sun, 20 Sep 2026 07:30:25 GMT<p>Jev picks Pokémon Red goals; code presses buttons.</p><p>A small autonomous agent that plays Pokémon Red on a PyBoy emulator. It is aimed at developers exploring TypeSafe Jev as a decision model for game or agent loops.</p><p><a href="https://github.com/milanboers/jev-plays-pokemon">milanboers/jev-plays-pokemon</a></p>Game & Simulationflat-fanouthysteresishttps://jevbooks.com/en/p/pithings__advocaathttps://jevbooks.com/en/p/pithings__advocaatType-safe client for asking Jev questions about your data.Sun, 20 Sep 2026 07:27:23 GMT<p>Type-safe client for asking Jev questions about your data.</p><p>Advocaat is a small TypeScript client for asking AI questions about a dataset, powered by TypeSafe Jev. It returns probabilities, choices, and scores in one request, aimed at developers building type-safe data triage and classification.</p><p><a href="https://github.com/pithings/advocaat">pithings/advocaat</a></p>Infra / SDKs / Integrationsflat-fanouthttps://jevbooks.com/en/p/r-ms__mini-jevhttps://jevbooks.com/en/p/r-ms__mini-jevReading option-letter logits instead of generating JSON on a frozen Qwen3-4B.Sun, 20 Sep 2026 07:27:13 GMT<p>Reading option-letter logits instead of generating JSON on a frozen Qwen3-4B.</p><p>A preregistered experiment and teaching bench measuring whether a frozen Qwen3-4B can answer closed-choice schema fields by reading next-token logits for option letters, instead of generating grammar-constrained JSON. Task: CLINC150 intent classification.</p><p><a href="https://github.com/r-ms/mini-jev">r-ms/mini-jev</a></p>Evaluation & Benchmarkinghttps://jevbooks.com/en/p/gaborishka__jev-canvashttps://jevbooks.com/en/p/gaborishka__jev-canvasVoice and finger drawing on tldraw canvas with Jev decisions.Sun, 20 Sep 2026 07:27:00 GMT<p>Voice and finger drawing on tldraw canvas with Jev decisions.</p><p>A multimodal UI where users draw on a tldraw canvas by speaking commands while pointing a webcam-tracked finger. It supports English and Ukrainian, and uses Jev to interpret spoken instructions into canvas actions.</p><p><a href="https://github.com/gaborishka/jev-canvas">gaborishka/jev-canvas</a></p>Agent Decisionsflat-fanouthttps://jevbooks.com/en/p/valentynkit__jev.nvimhttps://jevbooks.com/en/p/valentynkit__jev.nvimAsk the buffer a question; functions ranked by probability in quickfix.Sun, 20 Sep 2026 07:26:47 GMT<p>Ask the buffer a question; functions ranked by probability in quickfix.</p><p>A Neovim plugin that splits the buffer into functions with Treesitter, asks a plain-language question about each, and ranks the answers by probability in quickfix. For developers searching code by intent rather than pattern.</p><p><a href="https://github.com/valentynkit/jev.nvim">valentynkit/jev.nvim</a></p>Scoring & Rankingstate-minimalismhttps://jevbooks.com/en/p/nyarlathoteppppp__pi-heedhttps://jevbooks.com/en/p/nyarlathoteppppp__pi-heedRuntime constraints that check every side-effecting pi tool call.Sun, 20 Sep 2026 07:26:37 GMT<p>Runtime constraints that check every side-effecting pi tool call.</p><p>pi-heed is a runtime constraint layer for the pi coding agent. It checks every side-effecting tool call against what the user actually asked for, before the call runs, and blocks calls that would violate the user's stated policy.</p><p><a href="https://github.com/Nyarlathoteppppp/pi-heed">Nyarlathoteppppp/pi-heed</a></p>Verification & Guardrailshttps://jevbooks.com/en/p/codealive-ai__mastra-jev-moderationhttps://jevbooks.com/en/p/codealive-ai__mastra-jev-moderationMastra input moderation on Jev: one file, one typed request.Sun, 20 Sep 2026 07:24:07 GMT<p>Mastra input moderation on Jev: one file, one typed request.</p><p>A Mastra input processor that moderates agent messages with TypeSafe Jev instead of an LLM, for teams that want a fast, cheap, fail-open block gate with no text to parse.</p><p><a href="https://github.com/CodeAlive-AI/mastra-jev-moderation">CodeAlive-AI/mastra-jev-moderation</a></p>Content Moderationflat-fanoutfailure-mode-knobstate-minimalismhttps://jevbooks.com/en/p/bitnovus__jev-spam-evalhttps://jevbooks.com/en/p/bitnovus__jev-spam-evalZero-shot spam filtering with Jev Noul questions, benchmarked against TF-IDF.Sun, 20 Sep 2026 07:23:55 GMT<p>Zero-shot spam filtering with Jev Noul questions, benchmarked against TF-IDF.</p><p>Explores zero-shot email classification with TypeSafe's pretrained Jev model, comparing ham/spam/phishing decisions against TF-IDF logistic regression baselines on main, fresh, and recent phishing sets.</p><p><a href="https://github.com/bitnovus/jev-spam-eval">bitnovus/jev-spam-eval</a></p>Classification & Routinghttps://jevbooks.com/en/p/brainstormity__jev-moderation-bothttps://jevbooks.com/en/p/brainstormity__jev-moderation-botDiscord moderation bot using Jev to score spam, phishing, and social engineering.Sun, 20 Sep 2026 07:23:54 GMT<p>Discord moderation bot using Jev to score spam, phishing, and social engineering.</p><p>A Discord moderation bot for server admins and moderators. It analyzes incoming messages in real time, deletes spam and scam links, escalates offenses through a four-stage ladder, and profiles members based on their message history.</p><p><a href="https://github.com/brainstormity/Jev-Moderation-Bot">brainstormity/Jev-Moderation-Bot</a></p>Content Moderationtiered-thresholdshttps://jevbooks.com/en/p/aowang-ai__jev-tradehttps://jevbooks.com/en/p/aowang-ai__jev-tradeLive Jev trader on HyperliquidSun, 20 Sep 2026 07:23:49 GMT<p>Live Jev trader on Hyperliquid</p><p>A trading bot that uses Jev to decide long or short and open, close, or hold on Hyperliquid markets. Runs five isolated sleeves (BTC, ETH, SOL, DOGE, BNB), each with its own wallet and Jev call.</p><p><a href="https://github.com/aowang-ai/jev-trade">aowang-ai/jev-trade</a></p>Finance & Tradinghttps://jevbooks.com/en/p/sosopop__jev_stockhttps://jevbooks.com/en/p/sosopop__jev_stockExperimental Jev framework forecasting Hong Kong stock direction from market state.Sun, 20 Sep 2026 07:23:39 GMT<p>Experimental Jev framework forecasting Hong Kong stock direction from market state.</p><p>An experimental Python framework for forecasting short-term Hong Kong stock price direction. It targets researchers who want a disciplined reasoning layer over structured market data, not a proven trading model.</p><p><a href="https://github.com/sosopop/jev_stock">sosopop/jev_stock</a></p>Finance & Tradingcontract-validationhttps://jevbooks.com/en/p/unicodeveloper__jevockshttps://jevbooks.com/en/p/unicodeveloper__jevocksStock decision terminal using Jev for probabilistic 30-day outlook.Sun, 20 Sep 2026 07:23:29 GMT<p>Stock decision terminal using Jev for probabilistic 30-day outlook.</p><p>Jevinik is a stock decision terminal that retrieves live market evidence with Valyu and estimates whether a stock will trade higher in 30 days. It is for users who want a fast, structured probabilistic outlook on US and European stocks.</p><p><a href="https://github.com/unicodeveloper/jevocks">unicodeveloper/jevocks</a></p>Finance & Tradingflat-fanouthttps://jevbooks.com/en/p/valentynkit__jev-plays-pokemon-redhttps://jevbooks.com/en/p/valentynkit__jev-plays-pokemon-redPokemon Red on PyBoy: code owns the route, Jev picks at branches.Sun, 20 Sep 2026 07:23:24 GMT<p>Pokemon Red on PyBoy: code owns the route, Jev picks at branches.</p><p>A Pokemon Red agent on PyBoy for developers: deterministic Python handles routes, thresholds and arithmetic, while Jev is asked only at game branches to choose among legal actions.</p><p><a href="https://github.com/valentynkit/jev-plays-pokemon-red">valentynkit/jev-plays-pokemon-red</a></p>Game & Simulationmeasured-calibrationhttps://jevbooks.com/en/p/omnijev__playjevhttps://jevbooks.com/en/p/omnijev__playjev0.8B vision-language model plays ten browser games from raw pixelsSun, 20 Sep 2026 07:23:23 GMT<p>0.8B vision-language model plays ten browser games from raw pixels</p><p>PlayJev is a fine-tune of Qwen3.5-0.8B-Base that plays ten classic browser games directly from raw pixels, aimed at real-time GUI-game agents and Jev-style decision models.</p><p><a href="https://github.com/OmniJev/PlayJev">OmniJev/PlayJev</a></p>Game & Simulationhttps://jevbooks.com/en/p/kavehmz__typesafe-playgroundhttps://jevbooks.com/en/p/kavehmz__typesafe-playgroundInteractive experiments with TypeSafe Jev for routing and driving decisions.Sun, 20 Sep 2026 07:23:22 GMT<p>Interactive experiments with TypeSafe Jev for routing and driving decisions.</p><p>Small interactive demos for developers to see how TypeSafe Jev makes decisions: routing a support message and steering a car in a 3D world with visible inputs and judgments.</p><p><a href="https://github.com/kavehmz/typesafe-playground">kavehmz/typesafe-playground</a></p>Game & Simulationhttps://jevbooks.com/en/p/kxzk__typesafe-jev-drone-demohttps://jevbooks.com/en/p/kxzk__typesafe-jev-drone-demoThree.js drone simulator with Python backend and live TypeSafe Jev navigation.Sun, 20 Sep 2026 07:23:14 GMT<p>Three.js drone simulator with Python backend and live TypeSafe Jev navigation.</p><p>A live Three.js drone simulation with a Python/FastAPI backend where TypeSafe's jev-latest chooses each navigation action. It is for developers exploring typed AI decision-making in a simulated flight environment.</p><p><a href="https://github.com/kxzk/typesafe-jev-drone-demo">kxzk/typesafe-jev-drone-demo</a></p>Agent Decisionscross-checkcontract-validationhttps://jevbooks.com/en/p/phyous__tsai-schttps://jevbooks.com/en/p/phyous__tsai-scTypeSafe Jev plays original StarCraft shareware mission Strongarm.Sun, 20 Sep 2026 07:23:02 GMT<p>TypeSafe Jev plays original StarCraft shareware mission Strongarm.</p><p>A Python harness that drives the original 1998 StarCraft shareware executable inside BottleShip, letting Jev choose commands via ordinary mouse and keyboard input to complete the Strongarm combat mission.</p><p><a href="https://github.com/phyous/tsai-sc">phyous/tsai-sc</a></p>Game & Simulationhttps://jevbooks.com/en/p/romanslack__jev-dronehttps://jevbooks.com/en/p/romanslack__jev-droneCamera-only MuJoCo drone with Jev judgment model in the loopSun, 20 Sep 2026 07:22:54 GMT<p>Camera-only MuJoCo drone with Jev judgment model in the loop</p><p>An autonomous quadrotor flies a five-station obstacle course in MuJoCo using only its onboard camera. Classical CV and control stay in code; a small Jev model decides what each situation means. For robotics researchers exploring judgment models in live control loops.</p><p><a href="https://github.com/RomanSlack/jev-drone">RomanSlack/jev-drone</a></p>Agent Decisionsflat-fanouthttps://jevbooks.com/en/p/fhshaik__typesafe-mariohttps://jevbooks.com/en/p/fhshaik__typesafe-marioJev picks NES controller inputs for Super Mario Bros. from structured state.Sun, 20 Sep 2026 07:22:51 GMT<p>Jev picks NES controller inputs for Super Mario Bros. from structured state.</p><p>An experimental controller that lets TypeSafe's Jev model directly choose NES controller inputs for the original Super Mario Bros., using emulator telemetry and RAM rather than screenshots.</p><p><a href="https://github.com/fhshaik/typesafe-mario">fhshaik/typesafe-mario</a></p>Game & Simulationhttps://jevbooks.com/en/p/obie__ruby_decision_modelhttps://jevbooks.com/en/p/obie__ruby_decision_modelRuby client for decision models like Jev.Sun, 20 Sep 2026 07:22:49 GMT<p>Ruby client for decision models like Jev.</p><p>A Ruby gem that talks to decision models such as Typesafe Jev, so Ruby applications can ask typed questions about a state and get calibrated probabilities instead of generated text.</p><p><a href="https://github.com/obie/ruby_decision_model">obie/ruby_decision_model</a></p>Infra / SDKs / Integrationsflat-fanouthttps://jevbooks.com/en/p/dabit3__jev-experimentshttps://jevbooks.com/en/p/dabit3__jev-experiments22 latency-focused Jev demo apps built by DevinSun, 20 Sep 2026 07:22:42 GMT<p>22 latency-focused Jev demo apps built by Devin</p><p>A collection of 22 latency-focused demo applications built with Jev by Devin. Each app lives in its own top-level directory with its own README, TESTING.md and screenshots, spanning shell guards, log sentinels, instant search, reranking, and voice turn-taking.</p><p><a href="https://github.com/dabit3/jev-experiments">dabit3/jev-experiments</a></p>Scoring & Rankinghttps://jevbooks.com/en/p/edgardcham__hunchohttps://jevbooks.com/en/p/edgardcham__hunchoDecisions as code on System One models: typed questions, thresholds with hysteresis, nested decisions, journal, calibrationSun, 20 Sep 2026 07:22:40 GMT<p>Decisions as code on System One models: typed questions, thresholds with hysteresis, nested decisions, journal, calibration</p><p>huncho is a TypeScript SDK for developers to define decisions as code. It asks typed questions of a decision model, applies thresholds with hysteresis, nests decisions, journals every one, replays policy changes without inference, and calibrates against outcomes.</p><p><a href="https://github.com/edgardcham/huncho">edgardcham/huncho</a></p>Infra / SDKs / Integrationsflat-fanouttiered-thresholdshysteresiscode-owned-policymeasured-calibrationhttps://jevbooks.com/en/p/shitianfang__jev-usehttps://jevbooks.com/en/p/shitianfang__jev-useHand text-free agent steps to Jev as typed judgments.Sun, 20 Sep 2026 07:22:39 GMT<p>Hand text-free agent steps to Jev as typed judgments.</p><p>A Claude Code, Codex, and pi plugin plus library that routes agent steps needing no text output to Jev, TypeSafe's judgment model, so LLMs and Jev collaborate: Jev decides fast, the LLM writes.</p><p><a href="https://github.com/shitianfang/jev-use">shitianfang/jev-use</a></p>Agent Decisionsflat-fanouthttps://jevbooks.com/en/p/ariel-frischer__jevkithttps://jevbooks.com/en/p/ariel-frischer__jevkitRust CLI for TypeSafe Jev: typed decisions, offline linting.Sun, 20 Sep 2026 07:22:33 GMT<p>Rust CLI for TypeSafe Jev: typed decisions, offline linting.</p><p>jevkit is a Rust command-line tool for developers and AI agents to ask TypeSafe Jev typed questions about a passage and receive probabilities, labels, and scores as JSON.</p><p><a href="https://github.com/ariel-frischer/jevkit">ariel-frischer/jevkit</a></p>Infra / SDKs / Integrationsflat-fanouthttps://jevbooks.com/en/p/mgaitan__sqlite-jevhttps://jevbooks.com/en/p/mgaitan__sqlite-jevBatched natural-language judgments for SQLite via TypeSafe Jev.Sun, 20 Sep 2026 07:22:31 GMT<p>Batched natural-language judgments for SQLite via TypeSafe Jev.</p><p>A SQLite loadable C extension plus Python package that brings TypeSafe Jev to SQLite. It exposes Noul, Choice, and Score judgments as SQL scalar functions and batched virtual-table queries, aimed at analysts and developers doing natural-language classification or scoring over table data.</p><p><a href="https://github.com/mgaitan/sqlite-jev">mgaitan/sqlite-jev</a></p>Classification & Routinghttps://jevbooks.com/en/p/kylemclaren__jevqlhttps://jevbooks.com/en/p/kylemclaren__jevqlSemantic SQL for vanilla Postgres, powered by JevSun, 20 Sep 2026 07:22:27 GMT<p>Semantic SQL for vanilla Postgres, powered by Jev</p><p>A psql-shaped CLI and Go/TypeScript/Python SDKs that run plain SQL on vanilla Postgres (no extension), then ask Jev questions about each surviving row so clients can filter, sort and group semantically.</p><p><a href="https://github.com/kylemclaren/jevql">kylemclaren/jevql</a></p>Data Labeling & Curationhttps://jevbooks.com/en/p/yusukebe__hono-jev-routerhttps://jevbooks.com/en/p/yusukebe__hono-jev-routerSemantic router for Hono that routes requests by meaning with Jev.Sun, 20 Sep 2026 07:22:24 GMT<p>Semantic router for Hono that routes requests by meaning with Jev.</p><p>An experimental router for Hono where routes are described in plain words instead of method and path. Jev decides which description fits the incoming request. Aimed at developers experimenting with semantic routing.</p><p><a href="https://github.com/yusukebe/hono-jev-router">yusukebe/hono-jev-router</a></p>Classification & Routingflat-fanouthttps://jevbooks.com/en/p/cline__pluginshttps://jevbooks.com/en/p/cline__pluginsOfficial curated plugins for Cline CLI and extensions.Sun, 20 Sep 2026 07:22:21 GMT<p>Official curated plugins for Cline CLI and extensions.</p><p>A default collection of plugins for Cline CLI and extensions, installed by slug. Includes a Jev-driven browser plugin, jev-browser, that brings Jev into Cline as a first-class capability.</p><p><a href="https://github.com/cline/plugins">cline/plugins</a></p>Agent Decisionshttps://jevbooks.com/en/p/rajivkuriakose__typesafe-jev-exampleshttps://jevbooks.com/en/p/rajivkuriakose__typesafe-jev-examplesWorked Jev examples: ticket triage and re-ranking, runnable via OpenRouter.Sun, 20 Sep 2026 07:22:18 GMT<p>Worked Jev examples: ticket triage and re-ranking, runnable via OpenRouter.</p><p>Starter examples for developers trying TypeSafe's Jev System One model. Includes a ticket-triage example and a re-ranking example, each with sample data and a Makefile, runnable through OpenRouter without an early-access key.</p><p><a href="https://github.com/rajivkuriakose/typesafe-jev-examples">rajivkuriakose/typesafe-jev-examples</a></p>Classification & Routingflat-fanouthttps://jevbooks.com/en/p/dakdevs__decide-mcphttps://jevbooks.com/en/p/dakdevs__decide-mcpLocal MCP server delegating decisions to configurable models with percentage scores.Sun, 20 Sep 2026 07:22:15 GMT<p>Local MCP server delegating decisions to configurable models with percentage scores.</p><p>A local stdio MCP server for agents that need a decision made. Send a decision, context, and 2–64 choices; receive a recommended choice with percentages. Defaults to TypeSafe AI Jev via AI Gateway.</p><p><a href="https://github.com/dakdevs/decide-mcp">dakdevs/decide-mcp</a></p>Agent Decisionshttps://jevbooks.com/en/p/tumf__jev-clihttps://jevbooks.com/en/p/tumf__jev-cliDependency-free CLI and MCP server for TypeSafe Jev.Sun, 20 Sep 2026 07:22:07 GMT<p>Dependency-free CLI and MCP server for TypeSafe Jev.</p><p>jev-cli is a small Python CLI and stdio MCP server for TypeSafe Jev, aimed at developers who need fast classification or judgment in application code instead of generated prose.</p><p><a href="https://github.com/tumf/jev-cli">tumf/jev-cli</a></p>Infra / SDKs / Integrationsflat-fanouthttps://jevbooks.com/en/p/stumble__jev-gohttps://jevbooks.com/en/p/stumble__jev-goCommunity Go SDK for TypeSafe AI Jev / System One.Sun, 20 Sep 2026 07:22:05 GMT<p>Community Go SDK for TypeSafe AI Jev / System One.</p><p>A community-maintained Go client for TypeSafe AI's Jev / System One API, for Go developers who want to ask typed questions over state and read calibrated answers.</p><p><a href="https://github.com/Stumble/jev-go">Stumble/jev-go</a></p>Infra / SDKs / Integrationshttps://jevbooks.com/en/p/dannote__jevhttps://jevbooks.com/en/p/dannote__jevTypeSafe Jev for OTP: GenServer replies with Jev's answer for pattern matching.Sun, 20 Sep 2026 07:21:55 GMT<p>TypeSafe Jev for OTP: GenServer replies with Jev's answer for pattern matching.</p><p>An Elixir library that integrates TypeSafe Jev, a peer process, with OTP GenServers. It lets callers pattern match on Jev's answers directly, enabling non-blocking, concurrent decision-making in Elixir applications.</p><p><a href="https://github.com/dannote/jev">dannote/jev</a></p>Classification & Routingflat-fanouttiered-thresholdshttps://jevbooks.com/en/p/tacticocc__jevbridgehttps://jevbooks.com/en/p/tacticocc__jevbridgeACP/MCP adapter bridging TypeSafe Jev typed decisions with any LLM.Sun, 20 Sep 2026 07:21:45 GMT<p>ACP/MCP adapter bridging TypeSafe Jev typed decisions with any LLM.</p><p>Jevbridge is a TypeScript adapter that exposes TypeSafe Jev's typed probabilistic decisions (noul, choice, score) to LLMs like Codex, Claude, Grok, and OpenCode via MCP and ACP, enabling routing, gating, scoring, and computer-use action selection.</p><p><a href="https://github.com/tacticocc/Jevbridge">tacticocc/Jevbridge</a></p>Agent Decisionsflat-fanouttiered-thresholdshttps://jevbooks.com/en/p/kitze__skillboxhttps://jevbooks.com/en/p/kitze__skillboxSelf-hosted versioned skills library for AI agents with optional Jev recommendations.Sun, 20 Sep 2026 07:21:43 GMT<p>Self-hosted versioned skills library for AI agents with optional Jev recommendations.</p><p>Skillbox is a self-hosted, versioned skills library for AI agents. It provides a Markdown/file editor, immutable revisions, profiles with grants, revocable client keys, HTTP MCP, and optional Jev recommendations using your own provider key.</p><p><a href="https://github.com/kitze/skillbox">kitze/skillbox</a></p>Agent Decisionshttps://jevbooks.com/en/p/smithersai__smithershttps://jevbooks.com/en/p/smithersai__smithersAgentic workflow framework for TypeScript flows that instrument a code repository.Sun, 20 Sep 2026 07:21:42 GMT<p>Agentic workflow framework for TypeScript flows that instrument a code repository.</p><p>Smithers is an agentic workflow framework from smithersai for defining workflows in TypeScript configuration files and executing them quickly, durably, and reliably. It targets maintainers who want repository changes to get cheaper, faster, and smarter.</p><p><a href="https://github.com/smithersai/smithers">smithersai/smithers</a></p>Verification & Guardrailshttps://jevbooks.com/en/p/typesafe-ai__skillshttps://jevbooks.com/en/p/typesafe-ai__skillsAgent skills for building with TypeSafe's System One APISun, 20 Sep 2026 07:21:37 GMT<p>Agent skills for building with TypeSafe's System One API</p><p>An installable agent skills package for developers building with TypeSafe's System One API, which provides typed decisions and probabilities. It teaches agents the Jev workflow for designing TypeSafe workflows and composing typed judgments in code.</p><p><a href="https://github.com/typesafe-ai/skills">typesafe-ai/skills</a></p>Agent Decisionshttps://jevbooks.com/en/p/andrelandgraf__safer-with-jevhttps://jevbooks.com/en/p/andrelandgraf__safer-with-jevNeon Function proxy that routes AI Gateway decisions with TypeSafe Jev.Sun, 20 Sep 2026 07:21:36 GMT<p>Neon Function proxy that routes AI Gateway decisions with TypeSafe Jev.</p><p>A public showcase of TypeSafe Jev judgments: a Neon Function proxy in front of the Neon AI Gateway. It inspects request bodies and optionally forwards the same bytes to a caller-chosen HTTPS URL. Aimed at developers and agents.</p><p><a href="https://github.com/andrelandgraf/safer-with-jev">andrelandgraf/safer-with-jev</a></p>Classification & Routinghttps://jevbooks.com/en/p/wiktorb2004__llama-index-jevhttps://jevbooks.com/en/p/wiktorb2004__llama-index-jevLlamaIndex reranker and router using TypeSafe Jev typed decisions.Sun, 20 Sep 2026 07:21:35 GMT<p>LlamaIndex reranker and router using TypeSafe Jev typed decisions.</p><p>An unofficial LlamaIndex adapter providing a reranker (JevRerank) and router selectors (JevSingleSelector, JevMultiSelector) powered by TypeSafe Jev, for RAG retrieval and tool selection.</p><p><a href="https://github.com/WiktorB2004/llama-index-jev">WiktorB2004/llama-index-jev</a></p>Scoring & Rankinghttps://jevbooks.com/en/p/butochnikov__laravel-typesafe-jevhttps://jevbooks.com/en/p/butochnikov__laravel-typesafe-jevUnofficial Laravel integration for TypeSafe Jev with typed responses and testing fakes.Sun, 20 Sep 2026 07:21:29 GMT<p>Unofficial Laravel integration for TypeSafe Jev with typed responses and testing fakes.</p><p>A community-maintained Laravel package for using TypeSafe Jev AI. It provides typed responses, async requests, scoped dependency injection, configuration, a facade, and a recording fake for testing, without adding text-generation or chat features.</p><p><a href="https://github.com/Butochnikov/laravel-typesafe-jev">Butochnikov/laravel-typesafe-jev</a></p>Infra / SDKs / Integrationsflat-fanouthttps://jevbooks.com/en/p/alterhq__typesafe-sdk-swifthttps://jevbooks.com/en/p/alterhq__typesafe-sdk-swiftUnofficial Swift 6 client for the TypeSafe AI APISun, 20 Sep 2026 07:21:29 GMT<p>Unofficial Swift 6 client for the TypeSafe AI API</p><p>A dependency-free Swift 6 client for the TypeSafe AI API, ported from the official JavaScript and Python SDKs. It targets macOS 13+, iOS/iPadOS 16+, tvOS 16+, and watchOS 9+, and is intended for developers who want to call Jev from Apple platforms.</p><p><a href="https://github.com/alterhq/typesafe-sdk-swift">alterhq/typesafe-sdk-swift</a></p>Infra / SDKs / Integrationsflat-fanouthttps://jevbooks.com/en/p/jamesward__zio-typesafe-aihttps://jevbooks.com/en/p/jamesward__zio-typesafe-aiScala 3 / ZIO client for TypeSafe AI's Jev typed-question API.Sun, 20 Sep 2026 07:21:21 GMT<p>Scala 3 / ZIO client for TypeSafe AI's Jev typed-question API.</p><p>A Scala 3 / ZIO library for TypeSafe AI's Jev / System One API, which answers typed atomic questions about a piece of state instead of generating text. For Scala developers who want typed, end-to-end access to Jev decisions.</p><p><a href="https://github.com/jamesward/zio-typesafe-ai">jamesward/zio-typesafe-ai</a></p>Infra / SDKs / Integrationsflat-fanouthttps://jevbooks.com/en/p/blakestone-x__jev-mcphttps://jevbooks.com/en/p/blakestone-x__jev-mcpMCP server exposing Jev typed judgments as tools for any agent.Sun, 20 Sep 2026 07:21:19 GMT<p>MCP server exposing Jev typed judgments as tools for any agent.</p><p>A Python MCP server that exposes TypeSafe Jev's typed classify, score, check, match, and screen judgments as tools, so any MCP-capable agent can get calibrated confidence on closed-answer decisions.</p><p><a href="https://github.com/blakestone-x/jev-mcp">blakestone-x/jev-mcp</a></p>Infra / SDKs / Integrationshttps://jevbooks.com/en/p/jkudish__jev-mcphttps://jevbooks.com/en/p/jkudish__jev-mcpTyped Jev judgments as MCP tools for agents.Sun, 20 Sep 2026 07:20:58 GMT<p>Typed Jev judgments as MCP tools for agents.</p><p>An MCP server exposing ten Jev judgment tools for agents: claim verification, content screening, candidate ranking, classification, comparison, extraction, diff review, and gating.</p><p><a href="https://github.com/jkudish/jev-mcp">jkudish/jev-mcp</a></p>Infra / SDKs / Integrationshttps://jevbooks.com/en/p/vercel-labs__ai-clihttps://jevbooks.com/en/p/vercel-labs__ai-cliVercel Labs CLI for generation and typed evaluation with Jev.Sun, 20 Sep 2026 07:20:53 GMT<p>Vercel Labs CLI for generation and typed evaluation with Jev.</p><p>ai-cli is a terminal tool from Vercel Labs that generates text, images, video, and audio, and evaluates typed questions. It is for developers who want composable AI commands with stdin support and predictable outputs.</p><p><a href="https://github.com/vercel-labs/ai-cli">vercel-labs/ai-cli</a></p>Evaluation & Benchmarkingflat-fanoutcode-owned-policyhttps://jevbooks.com/en/p/wfzyx__vonhttps://jevbooks.com/en/p/wfzyx__vonOpen-source non-autoregressive System One decision model for calibrated inference.Sun, 20 Sep 2026 07:20:40 GMT<p>Open-source non-autoregressive System One decision model for calibrated inference.</p><p>Von is a 395M-parameter non-autoregressive model that answers typed questions over input state with calibrated probabilities in sub-25ms, for developers needing local, deterministic decision-making without generative text.</p><p><a href="https://github.com/wfzyx/von">wfzyx/von</a></p>Classification & Routingmeasured-calibrationhttps://jevbooks.com/en/p/david-cermak__jevlike-esp32https://jevbooks.com/en/p/david-cermak__jevlike-esp32Jevlike edge router on ESP32Sun, 20 Sep 2026 07:20:39 GMT<p>Jevlike edge router on ESP32</p><p>A demo that exports a jevlike text model scorer as ESP-IDF firmware for ESP32, with training in Python and a host-side check. For embedded developers exploring jevlike AI on microcontrollers.</p><p><a href="https://github.com/david-cermak/jevlike-esp32">david-cermak/jevlike-esp32</a></p>Classification & Routinghttps://jevbooks.com/en/p/olanotolu__jevbetterhttps://jevbooks.com/en/p/olanotolu__jevbetterOne-pass scorer over changing text options, stronger than jevlike.Sun, 20 Sep 2026 07:20:33 GMT<p>One-pass scorer over changing text options, stronger than jevlike.</p><p>jevbetter is a from-scratch Python one-pass option scorer: given a context and a variable list of text options, it returns one probability per option. It targets routing, ranking, open-label classification, and game controllers.</p><p><a href="https://github.com/olanotolu/jevbetter">olanotolu/jevbetter</a></p>Scoring & Rankingmeasured-calibrationhttps://jevbooks.com/en/p/vinnylarouge__jevlikehttps://jevbooks.com/en/p/vinnylarouge__jevlikeTrain a small model to score text options in one pass.Sun, 20 Sep 2026 07:20:33 GMT<p>Train a small model to score text options in one pass.</p><p>A training library for building a small model that takes a piece of text and a list of N text options, returning one probability per option in a single pass. For developers exploring Jev-like scoring.</p><p><a href="https://github.com/vinnylarouge/jevlike">vinnylarouge/jevlike</a></p>Scoring & Rankinghttps://jevbooks.com/en/p/zhengxuyu__litjevhttps://jevbooks.com/en/p/zhengxuyu__litjevReproduction of Jev: turn any Qwen model into a fast decision model.Sun, 20 Sep 2026 07:20:32 GMT<p>Reproduction of Jev: turn any Qwen model into a fast decision model.</p><p>LitJev is an independent research reproduction of the Jev decision layer on top of off-the-shelf Hugging Face Qwen checkpoints. Define questions and options, load a Qwen model, and get typed choices with probability distributions from one API call, with no training and no generated answer text.</p><p><a href="https://github.com/zhengxuyu/litjev">zhengxuyu/litjev</a></p>Scoring & Rankingflat-fanouthttps://jevbooks.com/en/p/us__jev-localhttps://jevbooks.com/en/p/us__jev-localLocal Jev-compatible evaluation server with typed questions and open weights.Sun, 20 Sep 2026 07:20:27 GMT<p>Local Jev-compatible evaluation server with typed questions and open weights.</p><p>A local server that answers typed Choice/Score/Noul questions with probabilities and confidence, compatible with the official Jev SDK. For developers who want to test Jev-style decisions without a waitlist or API key.</p><p><a href="https://github.com/us/jev-local">us/jev-local</a></p>Evaluation & Benchmarkingflat-fanouthttps://jevbooks.com/en/p/achimala__jev-painthttps://jevbooks.com/en/p/achimala__jev-paintPaint images by having Jev predict each pixel's colour.Sun, 20 Sep 2026 07:20:18 GMT<p>Paint images by having Jev predict each pixel's colour.</p><p>A local browser app that turns Jev's pixel probability distributions into paintings. Aimed at people with a TypeSafe Jev API key who want to run a small creative experiment on their own machine.</p><p><a href="https://github.com/achimala/jev-paint">achimala/jev-paint</a></p>Game & Simulationhttps://jevbooks.com/en/p/jaredpalmer__kevhttps://jevbooks.com/en/p/jaredpalmer__kevTrainable Jev-like decision models on Qwen3.5 you can run yourself.Sun, 20 Sep 2026 07:20:18 GMT<p>Trainable Jev-like decision models on Qwen3.5 you can run yourself.</p><p>Kev is a family of small decision models (0.8B, 4B, 9B) built on Qwen3.5, with training code, evaluation data, and a web playground. It's for developers who want Jev-style typed decisions locally.</p><p><a href="https://github.com/jaredpalmer/kev">jaredpalmer/kev</a></p>Agent Decisionsflat-fanouthttps://jevbooks.com/en/p/ikermoel__open-alternative-jevhttps://jevbooks.com/en/p/ikermoel__open-alternative-jevOpen-source System One model: typed, calibrated decisions from any open-weights LLM.Sun, 20 Sep 2026 07:20:10 GMT<p>Open-source System One model: typed, calibrated decisions from any open-weights LLM.</p><p>A Python package that gives typed, calibrated decisions from any open-weights LLM in one forward pass, running locally on your own GPU. For developers who want Jev-style decision-making without the official SDK.</p><p><a href="https://github.com/ikermoel/open-alternative-jev">ikermoel/open-alternative-jev</a></p>Classification & Routingflat-fanoutmeasured-calibrationhttps://jevbooks.com/en/p/tianyucodings__nanojevhttps://jevbooks.com/en/p/tianyucodings__nanojevA 0.6B parallel decision model returning full probability distributions.Sun, 20 Sep 2026 07:20:07 GMT<p>A 0.6B parallel decision model returning full probability distributions.</p><p>NanoJev is an open replica of Jev: a 0.6B parallel decision model that takes states and questions and returns complete probability distributions without output-token decoding, shipped with its training pipeline, weights, and dataset for four game tasks.</p><p><a href="https://github.com/TianyuCodings/NanoJev">TianyuCodings/NanoJev</a></p>Game & Simulationhttps://jevbooks.com/en/p/zhihz__openjevhttps://jevbooks.com/en/p/zhihz__openjevLocal bilingual probability decisions from context, questions, and candidates.Sun, 20 Sep 2026 07:20:01 GMT<p>Local bilingual probability decisions from context, questions, and candidates.</p><p>Open JEV is an independent research preview that answers bilingual probability questions from context, questions, and candidate answers. It is aimed at developers who need a local decision interface for applications.</p><p><a href="https://github.com/zhihz/openjev">zhihz/openjev</a></p>Scoring & Rankingflat-fanouthttps://jevbooks.com/en/p/mapika__deciderhttps://jevbooks.com/en/p/mapika__deciderOne-pass typed decisions with calibrated probabilities, no text generation.Sun, 20 Sep 2026 07:19:59 GMT<p>One-pass typed decisions with calibrated probabilities, no text generation.</p><p>decider is an open reproduction of the System One model class (TypeSafe AI's Jev). It reads a state and typed questions, returning calibrated probability distributions for each question in one forward pass. Released as 2B and 35B MoE models, for developers needing fast, structured decisions.</p><p><a href="https://github.com/Mapika/decider">Mapika/decider</a></p>Agent Decisionsflat-fanouthttps://jevbooks.com/en/p/scienthoon__jev-ood-calibrationhttps://jevbooks.com/en/p/scienthoon__jev-ood-calibrationIndependent calibration test of Jev on unseen rule-based tasks.Sun, 20 Sep 2026 07:19:55 GMT<p>Independent calibration test of Jev on unseen rule-based tasks.</p><p>An independent calibration study of TypeSafe's Jev, measuring whether its probabilities stay honest on inputs it cannot have seen: 900 rule-generated support tickets plus three public benchmarks, with every raw response published.</p><p><a href="https://github.com/scienthoon/jev-ood-calibration">scienthoon/jev-ood-calibration</a></p>Calibration & Researchhttps://jevbooks.com/en/p/yodablocks__jev-orderby-benchhttps://jevbooks.com/en/p/yodablocks__jev-orderby-benchIndependent measurement of whether ORDER BY over Jev probabilities is defensible.Sun, 20 Sep 2026 07:19:52 GMT<p>Independent measurement of whether ORDER BY over Jev probabilities is defensible.</p><p>A benchmark harness measuring whether SQL ORDER BY over Jev probabilities yields a defensible order, for developers building semantic sort on Jev. It reports pairwise inversion, Score ordinality, calibration, wording invariants and sort-key ties against pre-registered gates.</p><p><a href="https://github.com/yodablocks/jev-orderby-bench">yodablocks/jev-orderby-bench</a></p>Scoring & Rankinghttps://jevbooks.com/en/p/shogo-nfrealmusic__jev-evalhttps://jevbooks.com/en/p/shogo-nfrealmusic__jev-evalThird-party check: Jev vs GPT-4o-mini and Claude Sonnet 4.5 on booking inquiry routing.Sun, 20 Sep 2026 07:19:49 GMT<p>Third-party check: Jev vs GPT-4o-mini and Claude Sonnet 4.5 on booking inquiry routing.</p><p>Compares Jev, TypeSafe AI's judgment-only model, against GPT-4o-mini and Claude Sonnet 4.5 under identical conditions on routing 60 synthetic booking inquiries for a photo-shoot service in Japan.</p><p><a href="https://github.com/Shogo-nfrealmusic/jev-eval">Shogo-nfrealmusic/jev-eval</a></p>Evaluation & Benchmarkingmeasured-calibrationhttps://jevbooks.com/en/p/abhixhek__jevcalhttps://jevbooks.com/en/p/abhixhek__jevcalCalibrate, threshold, and drift-check typed decision models against an LLM teacher.Sun, 20 Sep 2026 07:19:46 GMT<p>Calibrate, threshold, and drift-check typed decision models against an LLM teacher.</p><p>jevcal measures a typed decision model on your labeled data, picks per-question confidence thresholds to meet your accuracy target, reports escalation traffic, and fails CI on drift. Built for Jev and any model returning answers with probabilities.</p><p><a href="https://github.com/abhixhek/jevcal">abhixhek/jevcal</a></p>Evaluation & Benchmarkingscreen-then-deephttps://jevbooks.com/en/p/ably-labs__jev-ponghttps://jevbooks.com/en/p/ably-labs__jev-pongPong where the ball advances one step per model decision.Sun, 20 Sep 2026 07:19:34 GMT<p>Pong where the ball advances one step per model decision.</p><p>A latency demo comparing Jev, a typed-decision model, against chat LLMs on Vercel AI Gateway. Each model plays the right paddle in Pong; one decision moves the ball one segment. Built by Ably's co-founder.</p><p><a href="https://github.com/ably-labs/jev-pong">ably-labs/jev-pong</a></p>Game & Simulationhttps://jevbooks.com/en/p/jgridifier__jev-research-evalhttps://jevbooks.com/en/p/jgridifier__jev-research-evalReproducible eval harness for Jev Ultrafast research-browser tasks.Sun, 20 Sep 2026 07:19:25 GMT<p>Reproducible eval harness for Jev Ultrafast research-browser tasks.</p><p>A reproducible evaluation harness for Jared's Jev Ultrafast research-browser session, with baseline cases, stress suites, QC grades, field note, and notebooks. For researchers evaluating browser agents.</p><p><a href="https://github.com/jgridifier/jev-research-eval">jgridifier/jev-research-eval</a></p>Evaluation & Benchmarkinghttps://jevbooks.com/en/p/hegargarcia__jev-playgroundhttps://jevbooks.com/en/p/hegargarcia__jev-playgroundBenchmarking Jev against other models in explicit-state games.Sun, 20 Sep 2026 07:19:25 GMT<p>Benchmarking Jev against other models in explicit-state games.</p><p>A Next.js playground for benchmarking TypeSafe AI's Jev against Luna, Haiku, Gemini, and Astra at choosing validated legal moves in tic-tac-toe and Connect Four, scoring decision quality and consistency across a sequence of moves.</p><p><a href="https://github.com/hegargarcia/jev-playground">hegargarcia/jev-playground</a></p>Evaluation & Benchmarkinghttps://jevbooks.com/en/p/cephalization__jev-triagehttps://jevbooks.com/en/p/cephalization__jev-triageMultiplayer triage dashboard where Jev answers typed questions per issue.Sun, 20 Sep 2026 07:19:23 GMT<p>Multiplayer triage dashboard where Jev answers typed questions per issue.</p><p>A multiplayer triage dashboard for public GitHub repositories. Issues and pull requests sync into Postgres, replicate live to every browser via Rocicorp Zero, and a TypeSafe System One model answers typed questions about each one.</p><p><a href="https://github.com/cephalization/jev-triage">cephalization/jev-triage</a></p>Classification & Routinghttps://jevbooks.com/en/p/akashpriyadarshii__jev-curatehttps://jevbooks.com/en/p/akashpriyadarshii__jev-curateRust sifter for synthetic and pretraining datasets using Jev.Sun, 20 Sep 2026 07:19:17 GMT<p>Rust sifter for synthetic and pretraining datasets using Jev.</p><p>jev-curate streams Parquet and JSONL datasets through TypeSafe AI's Jev model to filter synthetic reasoning data, instruction tuning pairs, and web-scraped corpora, emitting clean and rejected records verbatim for dataset engineers.</p><p><a href="https://github.com/AkashPriyadarshii/jev-curate">AkashPriyadarshii/jev-curate</a></p>Data Labeling & Curationhttps://jevbooks.com/en/p/yikangy873-gif__jev-desktophttps://jevbooks.com/en/p/yikangy873-gif__jev-desktopBounded Jev action selection inside Codex Computer UseSun, 20 Sep 2026 07:19:14 GMT<p>Bounded Jev action selection inside Codex Computer Use</p><p>A Codex plugin that adds a bounded decision loop to Codex Computer Use for browser tabs and native macOS apps, letting Codex delegate per-step action selection to TypeSafe Jev.</p><p><a href="https://github.com/yikangy873-gif/jev-desktop">yikangy873-gif/jev-desktop</a></p>Agent Decisionscontract-validationcandidate-enumerationhttps://jevbooks.com/en/p/tamaratran__jev-prunerhttps://jevbooks.com/en/p/tamaratran__jev-prunerClaude Code plugin trims long Bash output with Jev before the model sees it.Sun, 20 Sep 2026 07:19:12 GMT<p>Claude Code plugin trims long Bash output with Jev before the model sees it.</p><p>A Claude Code plugin for developers that uses TypeSafe's Jev to prune noisy Bash stdout after a command runs but before the result reaches the main LLM, reducing tokens carried into later turns without generating a summary.</p><p><a href="https://github.com/tamaratran/jev-pruner">tamaratran/jev-pruner</a></p>Infra / SDKs / Integrationsflat-fanouthttps://jevbooks.com/en/p/chy4pro__jev-for-chromehttps://jevbooks.com/en/p/chy4pro__jev-for-chromeChrome extension driving your tab with TypeSafe Jev decisions.Sun, 20 Sep 2026 07:19:10 GMT<p>Chrome extension driving your tab with TypeSafe Jev decisions.</p><p>A Manifest V3 Chrome extension that automates the tab you are looking at, using your own cookies and logins. It is a community port of browser-use/jev-ultrafast, not affiliated with TypeSafe or Browser Use.</p><p><a href="https://github.com/chy4pro/jev-for-chrome">chy4pro/jev-for-chrome</a></p>Agent Decisionsflat-fanoutcandidate-enumerationhttps://jevbooks.com/en/p/valentynkit__jev-belayhttps://jevbooks.com/en/p/valentynkit__jev-belayClaude Code Stop hook that blocks unverified "done" claims.Sun, 20 Sep 2026 07:19:07 GMT<p>Claude Code Stop hook that blocks unverified &quot;done&quot; claims.</p><p>A Claude Code Stop hook for coding agents: it reads the turn's transcript locally and blocks a turn that claims done when files changed and no check passed since.</p><p><a href="https://github.com/valentynkit/jev-belay">valentynkit/jev-belay</a></p>Verification & Guardrailsflat-fanoutfailure-mode-knobmeasured-calibrationhttps://jevbooks.com/en/p/goldenloaf24h__browserclawhttps://jevbooks.com/en/p/goldenloaf24h__browserclawChrome automation MCP server with local Jev semantic micro-loop.Sun, 20 Sep 2026 07:18:58 GMT<p>Chrome automation MCP server with local Jev semantic micro-loop.</p><p>BrowserClaw is a Chrome browser automation MCP server and agent skill for AI agents. It runs inside your everyday Chrome via an MV3 extension and Native Messaging bridge, providing 48 tools for navigation, DOM perception, and interaction without losing logins or stealing focus.</p><p><a href="https://github.com/GoldenLoaf24h/browserclaw">GoldenLoaf24h/browserclaw</a></p>Agent Decisionsscreen-then-deephttps://jevbooks.com/en/p/shitianfang__wakegatehttps://jevbooks.com/en/p/shitianfang__wakegateAsk Jev if a sleeping agent's wakeup is worth an LLM turn.Sun, 20 Sep 2026 07:18:58 GMT<p>Ask Jev if a sleeping agent's wakeup is worth an LLM turn.</p><p>A fail-open wake gate for long-running agents on Workers, Durable Objects and Node. Before a sleeping agent's LLM is resumed on a timer or incoming event, wakegate asks Jev whether the wakeup is worth a full LLM turn.</p><p><a href="https://github.com/shitianfang/wakegate">shitianfang/wakegate</a></p>Agent Decisionshysteresishttps://jevbooks.com/en/p/jekozyra__pi-typesafe-routerhttps://jevbooks.com/en/p/jekozyra__pi-typesafe-routerTyped Jev decisions route Pi coding-agent requests to the right model.Sun, 20 Sep 2026 07:18:57 GMT<p>Typed Jev decisions route Pi coding-agent requests to the right model.</p><p>A Pi extension for coding agents that classifies incoming requests with Jev and routes each one to a configured provider/model. It ships config examples for TypeSafe, Cloudflare, Vercel, and OpenRouter, and is meant to be the only automatic router enabled.</p><p><a href="https://github.com/jekozyra/pi-typesafe-router">jekozyra/pi-typesafe-router</a></p>Classification & Routinghttps://jevbooks.com/en/p/leonaaardob__fast-dev-compactionhttps://jevbooks.com/en/p/leonaaardob__fast-dev-compactionCodex plugin: verbatim Jev-guided context restoration around session compaction.Sun, 20 Sep 2026 07:18:50 GMT<p>Codex plugin: verbatim Jev-guided context restoration around session compaction.</p><p>An experimental Codex plugin and npm library that wraps Codex's built-in compaction. It prunes tool calls and results with Jev decisions, then re-injects the verbatim kept history as additionalContext. The README warns against using it in real workflows.</p><p><a href="https://github.com/leonaaardob/fast-dev-compaction">leonaaardob/fast-dev-compaction</a></p>Scoring & Rankingflat-fanouthttps://jevbooks.com/en/p/tamaratran__fast-jev-compactionhttps://jevbooks.com/en/p/tamaratran__fast-jev-compactionClaude Code plugin that prunes context with Jev instead of summarizing.Sun, 20 Sep 2026 07:18:49 GMT<p>Claude Code plugin that prunes context with Jev instead of summarizing.</p><p>An npm library and Claude Code plugin that replaces Claude Code's built-in compaction summary. It deletes tool calls and results Jev says are no longer needed, keeping all user and assistant text verbatim and in order.</p><p><a href="https://github.com/tamaratran/fast-jev-compaction">tamaratran/fast-jev-compaction</a></p>Agent Decisionshttps://jevbooks.com/en/p/bastani-inc__atomichttps://jevbooks.com/en/p/bastani-inc__atomicVerifiable coding agent runtime with typed Jev decision provider.Sun, 20 Sep 2026 07:18:48 GMT<p>Verifiable coding agent runtime with typed Jev decision provider.</p><p>Atomic is a TypeScript coding agent runtime for engineers who want their process expressed as explicit, checkable execution graphs with stages, checks, and approval gates, prioritizing verified production-ready work over quick responses.</p><p><a href="https://github.com/bastani-inc/atomic">bastani-inc/atomic</a></p>Agent Decisionshttps://jevbooks.com/en/p/joelhooks__pi-fast-jev-compactionhttps://jevbooks.com/en/p/joelhooks__pi-fast-jev-compactionPi extension pruning stale tool history with TypeSafe Jev decisions.Sun, 20 Sep 2026 07:18:42 GMT<p>Pi extension pruning stale tool history with TypeSafe Jev decisions.</p><p>A stock Pi 0.85.1 extension for developers managing long agent sessions. It keeps conversation text verbatim while pruning stale tool history, falling back to Pi's built-in summary compaction only when pruning cannot free enough context.</p><p><a href="https://github.com/joelhooks/pi-fast-jev-compaction">joelhooks/pi-fast-jev-compaction</a></p>Agent Decisionshttps://jevbooks.com/en/p/jkudish__jev-browserhttps://jevbooks.com/en/p/jkudish__jev-browserFast, cheap browser automation with Jev picking each step.Sun, 20 Sep 2026 07:18:42 GMT<p>Fast, cheap browser automation with Jev picking each step.</p><p>jev-browser drives a real headless browser from a task and a start URL, via an MCP server, CLI, or library. It is aimed at developers who want browser automation that is fast and very cheap compared to LLM-driven browsing.</p><p><a href="https://github.com/jkudish/jev-browser">jkudish/jev-browser</a></p>Agent Decisionscross-checkhttps://jevbooks.com/en/p/akashpriyadarshii__jev-superpowershttps://jevbooks.com/en/p/akashpriyadarshii__jev-superpowersSystematic agentic skills with TypeSafe Jev typed decisions.Sun, 20 Sep 2026 07:18:33 GMT<p>Systematic agentic skills with TypeSafe Jev typed decisions.</p><p>A software development framework for AI coding agents, upgrading the obra/superpowers methodology with TypeSafe Jev System One to gate architecture, dependency, shell, and completion decisions.</p><p><a href="https://github.com/AkashPriyadarshii/jev-superpowers">AkashPriyadarshii/jev-superpowers</a></p>Agent Decisionshttps://jevbooks.com/en/p/kevthetech143__super-jevhttps://jevbooks.com/en/p/kevthetech143__super-jevDecision harness: turns Jev answers into bounded actions.Sun, 20 Sep 2026 07:18:24 GMT<p>Decision harness: turns Jev answers into bounded actions.</p><p>A small, extensible decision-to-action harness for TypeSafe Jev. It connects evidence, Jev judgments, permitted actions, and verified outcomes for one bounded run. Domain-independent core with pluggable data sources and tools. For developers building agent-like loops with Jev.</p><p><a href="https://github.com/Kevthetech143/super-jev">Kevthetech143/super-jev</a></p>Verification & Guardrailsflat-fanouthttps://jevbooks.com/en/p/agent-labs-dev__fastbrowsehttps://jevbooks.com/en/p/agent-labs-dev__fastbrowseBrowser agent where Jev picks actions and an LLM plans and reads.Sun, 20 Sep 2026 07:18:20 GMT<p>Browser agent where Jev picks actions and an LLM plans and reads.</p><p>fastbrowse is an experimental browser agent for tasks like lookups, sign-ins and checkout. It indexes the page into candidates, has Jev pick one action, and returns answers where every claim cites a verbatim quote.</p><p><a href="https://github.com/agent-labs-dev/fastbrowse">agent-labs-dev/fastbrowse</a></p>Agent Decisionshttps://jevbooks.com/en/p/hyunjunjeon__pi-quiet-askhttps://jevbooks.com/en/p/hyunjunjeon__pi-quiet-askTypeSafe Jev as the pi coding agent's quiet decision layer.Sun, 20 Sep 2026 07:18:12 GMT<p>TypeSafe Jev as the pi coding agent's quiet decision layer.</p><p>A pi coding agent extension that answers closed, high-frequency harness questions with Jev, a System One model returning typed probabilities instead of text, so the main LLM keeps writing code.</p><p><a href="https://github.com/HyunjunJeon/pi-quiet-ask">HyunjunJeon/pi-quiet-ask</a></p>Verification & Guardrailsflat-fanouthttps://jevbooks.com/en/p/theooliveira__pi-jevhttps://jevbooks.com/en/p/theooliveira__pi-jevSemantic tool routing and typed System One decisions for Pi coding agent.Sun, 20 Sep 2026 07:18:09 GMT<p>Semantic tool routing and typed System One decisions for Pi coding agent.</p><p>A Pi coding agent extension that semantically routes tools and skills, and runs typed System One judgments using TypeSafe Jev. For developers using Pi who want fast, calibrated decisions without LLM overhead.</p><p><a href="https://github.com/TheoOliveira/pi-jev">TheoOliveira/pi-jev</a></p>Agent Decisionsfailure-mode-knobrules-as-questionshttps://jevbooks.com/en/p/compozy__yoshihttps://jevbooks.com/en/p/compozy__yoshiLocal context-pruning proxy for Claude Code and Codex.Sun, 20 Sep 2026 07:18:05 GMT<p>Local context-pruning proxy for Claude Code and Codex.</p><p>Yoshi is a local proxy that sits between coding agents (Claude Code, Codex) and their providers, judging which conversation history is still needed and pruning the rest before forwarding requests.</p><p><a href="https://github.com/compozy/yoshi">compozy/yoshi</a></p>Infra / SDKs / Integrationsfailure-mode-knobhttps://jevbooks.com/en/p/24601__augustushttps://jevbooks.com/en/p/24601__augustusDesign judgment for placing typed probabilistic judgment in software.Sun, 20 Sep 2026 07:18:00 GMT<p>Design judgment for placing typed probabilistic judgment in software.</p><p>Augustus is an agent skill for designing systems that use Jev-class typed probabilistic judgment (Choice, Score, Noul). It maps judgment onto classical decision methods and applies across AI, software, business, knowledge work, and life.</p><p><a href="https://github.com/24601/Augustus">24601/Augustus</a></p>Agent Decisionshttps://jevbooks.com/en/p/grmkris__robo-harnesshttps://jevbooks.com/en/p/grmkris__robo-harnessSO-101 robot-arm workbench with a Jev decision runner for bounded joint steps.Sun, 20 Sep 2026 07:17:47 GMT<p>SO-101 robot-arm workbench with a Jev decision runner for bounded joint steps.</p><p>A standalone SO-101 playground for developers: a Bun/Effect coordinator, React workbench, Rerun visual workspace, explicit robot tools, and a LeRobot hardware boundary, defaulting to mock hardware on a fresh checkout.</p><p><a href="https://github.com/grmkris/robo-harness">grmkris/robo-harness</a></p>Agent Decisionshttps://jevbooks.com/en/p/noplan-inc__limpethttps://jevbooks.com/en/p/noplan-inc__limpetStop hook that keeps coding agents from stopping too early.Sun, 20 Sep 2026 07:17:36 GMT<p>Stop hook that keeps coding agents from stopping too early.</p><p>limpet is a Stop hook for coding agents. You write completion rules in plain language; every time the agent is about to stop, limpet checks the stop against those rules and sends the agent back to work if a rule is violated.</p><p><a href="https://github.com/noplan-inc/limpet">noplan-inc/limpet</a></p>Agent Decisionsflat-fanouthttps://jevbooks.com/en/p/hyunjunjeon__jev-judgmenthttps://jevbooks.com/en/p/hyunjunjeon__jev-judgmentSend closed coding-agent judgments to TypeSafe JevSun, 20 Sep 2026 07:17:27 GMT<p>Send closed coding-agent judgments to TypeSafe Jev</p><p>An Agent Skill for coding agents that sends closed judgments to TypeSafe Jev and acts on the returned probabilities, so the main coding model keeps writing code and interrupts the user only when a person must decide.</p><p><a href="https://github.com/HyunjunJeon/jev-judgment">HyunjunJeon/jev-judgment</a></p>Agent Decisionsflat-fanoutfailure-mode-knobhttps://jevbooks.com/en/p/legacybridge-tech__pi-typesafe-jevhttps://jevbooks.com/en/p/legacybridge-tech__pi-typesafe-jevPi extension: TypeSafe Jev judgments as five pi tools.Sun, 20 Sep 2026 07:17:19 GMT<p>Pi extension: TypeSafe Jev judgments as five pi tools.</p><p>A pi extension that exposes TypeSafe (Jev, System One) judgments as five pi tools, letting a model make narrow semantic judgments while your code and users keep control of thresholds, weights, and actions.</p><p><a href="https://github.com/legacybridge-tech/pi-typesafe-jev">legacybridge-tech/pi-typesafe-jev</a></p>Infra / SDKs / Integrationsflat-fanouthttps://jevbooks.com/en/p/forvela__jev-agent-browserhttps://jevbooks.com/en/p/forvela__jev-agent-browserBounded browser agents: Jev picks typed actions, agent-browser executes.Sun, 20 Sep 2026 07:17:16 GMT<p>Bounded browser agents: Jev picks typed actions, agent-browser executes.</p><p>A delegated browser executor for parent agents. A parent gives a bounded task; Jev selects one typed browser action from the current snapshot, agent-browser runs it, and the loop returns success or a structured handoff.</p><p><a href="https://github.com/forvela/jev-agent-browser">forvela/jev-agent-browser</a></p>Agent Decisionshttps://jevbooks.com/en/p/browser-use__jev-ultrafasthttps://jevbooks.com/en/p/browser-use__jev-ultrafastBrowser agent where Jev picks each operation and element.Sun, 20 Sep 2026 07:17:06 GMT<p>Browser agent where Jev picks each operation and element.</p><p>A browser automation agent from browser-use. Given one natural-language goal, it drives a real Chrome page, choosing operations and target elements itself and calling a small LLM only when text must be typed.</p><p><a href="https://github.com/browser-use/jev-ultrafast">browser-use/jev-ultrafast</a></p>Agent Decisionscandidate-enumerationhttps://jevbooks.com/en/p/socai-io__jev-socialhttps://jevbooks.com/en/p/socai-io__jev-socialJev picks each social research step; socai runs it in your browser.Sun, 20 Sep 2026 07:16:56 GMT<p>Jev picks each social research step; socai runs it in your browser.</p><p>A browser-grounded research tool for Instagram, TikTok, and LinkedIn. It lets a user state a goal, then Jev chooses read-only socai operations while a real Chrome session captures posts, profiles, comments, and cited evidence.</p><p><a href="https://github.com/socai-io/jev-social">socai-io/jev-social</a></p>Agent Decisionshttps://jevbooks.com/en/p/valentynkit__jev-skiphttps://jevbooks.com/en/p/valentynkit__jev-skipYouTube sponsor skipper using Jev to score captions at watch time.Sun, 20 Sep 2026 07:16:46 GMT<p>YouTube sponsor skipper using Jev to score captions at watch time.</p><p>A browser extension that reads the YouTube caption track and scores each segment's sponsor probability on the seek bar before the intro ends, for viewers of videos nobody has labeled yet.</p><p><a href="https://github.com/valentynkit/jev-skip">valentynkit/jev-skip</a></p>Scoring & Rankinghttps://jevbooks.com/en/p/hotchpotch__jev-rerankerhttps://jevbooks.com/en/p/hotchpotch__jev-rerankerJev-powered relevance filtering and reranking for RAG in Python.Sun, 20 Sep 2026 07:16:40 GMT<p>Jev-powered relevance filtering and reranking for RAG in Python.</p><p>A Python library for RAG pipelines that reranks search results and filters retrieved documents by usefulness as answer evidence, using TypeSafe.AI's Jev. It is for developers who want to reduce context tokens and distracting passages before generation.</p><p><a href="https://github.com/hotchpotch/jev-reranker">hotchpotch/jev-reranker</a></p>Scoring & Rankinghttps://jevbooks.com/en/p/supercorp-ai__supercovhttps://jevbooks.com/en/p/supercorp-ai__supercovCode quality and coverage for coding agentsSun, 20 Sep 2026 07:16:39 GMT<p>Code quality and coverage for coding agents</p><p>Supercov scores code quality with Jev and measures test coverage, turning uncovered paths into actionable queries for coding agents to fix and test.</p><p><a href="https://github.com/supercorp-ai/supercov">supercorp-ai/supercov</a></p>Scoring & Rankinghttps://jevbooks.com/en/p/tky-27__jevslophttps://jevbooks.com/en/p/tky-27__jevslopScores note.com articles for AI slop using Jev.Sun, 20 Sep 2026 07:16:38 GMT<p>Scores note.com articles for AI slop using Jev.</p><p>JevSlop is a web app for readers of public note.com articles. Paste an article URL and it shows a transparent AI Slop Score plus eight writing-quality detail signals, with Japanese and English UI.</p><p><a href="https://github.com/TKY-27/JevSlop">TKY-27/JevSlop</a></p>Scoring & Rankingflat-fanouthttps://jevbooks.com/en/p/akashpriyadarshii__jev-seohttps://jevbooks.com/en/p/akashpriyadarshii__jev-seoZero-cost SEO & GEO radar CLI and MCP server for developers and agents.Sun, 20 Sep 2026 07:16:32 GMT<p>Zero-cost SEO &amp; GEO radar CLI and MCP server for developers and agents.</p><p>A free, open-source SEO and Generative Engine Optimization (GEO) search radar CLI suite and MCP server for developers and coding agents. It replaces paid tools like Semrush and OpenSEO by scraping DuckDuckGo locally and using TypeSafe Jev System One for deterministic evaluation.</p><p><a href="https://github.com/AkashPriyadarshii/jev-seo">AkashPriyadarshii/jev-seo</a></p>Infra / SDKs / Integrationshttps://jevbooks.com/en/p/akashpriyadarshii__jev-scouthttps://jevbooks.com/en/p/akashpriyadarshii__jev-scoutZero-hallucination repo and crate scout scored by TypeSafe AI Jev.Sun, 20 Sep 2026 07:16:30 GMT<p>Zero-hallucination repo and crate scout scored by TypeSafe AI Jev.</p><p>A Rust CLI and stdio MCP server that finds open-source repositories and crates from natural-language queries. Built for developers and autonomous coding agents that need verified, installable packages instead of hallucinated names.</p><p><a href="https://github.com/AkashPriyadarshii/jev-scout">AkashPriyadarshii/jev-scout</a></p>Scoring & Rankinghttps://jevbooks.com/en/p/kitze__pagegradehttps://jevbooks.com/en/p/kitze__pagegradeGrades page sections for clarity, writing and on-page SEO.Sun, 20 Sep 2026 07:16:29 GMT<p>Grades page sections for clarity, writing and on-page SEO.</p><p>PageGrade is a WXT Chrome extension (Manifest V3) for anyone who wants per-section feedback on a page's clarity, writing and on-page SEO. It shows ten rubric scores, an A–E grade and overlays badges on the page.</p><p><a href="https://github.com/kitze/pagegrade">kitze/pagegrade</a></p>Scoring & Rankinghttps://jevbooks.com/en/p/superagents-lab__jev-searchhttps://jevbooks.com/en/p/superagents-lab__jev-searchPlain-language web search ranked by Jev relevance judgments.Sun, 20 Sep 2026 07:16:23 GMT<p>Plain-language web search ranked by Jev relevance judgments.</p><p>Jev Search lets users search the open web in plain language. Jev chooses sources, time ranges and search terms, then ranks results from Search1API. It returns links and snippets with visible relevance scores and editable filters, and no generated answers.</p><p><a href="https://github.com/superagents-lab/jev-search">superagents-lab/jev-search</a></p>Scoring & Rankinghttps://jevbooks.com/en/p/komikat__jev-bfshttps://jevbooks.com/en/p/komikat__jev-bfsWikipedia link races ranked by Jev, shown live in your terminal.Sun, 20 Sep 2026 07:16:23 GMT<p>Wikipedia link races ranked by Jev, shown live in your terminal.</p><p>A Python CLI that finds a link path between two English Wikipedia articles. Chromium fetches pages, Jev ranks outgoing links, and Python controls a beam search across BFS levels. For people exploring Wikipedia connections.</p><p><a href="https://github.com/komikat/jev-bfs">komikat/jev-bfs</a></p>Scoring & Rankinghttps://jevbooks.com/en/p/marissafamularo__citation-verifierhttps://jevbooks.com/en/p/marissafamularo__citation-verifierCheck if cited papers support citing sentences.Sun, 20 Sep 2026 07:16:20 GMT<p>Check if cited papers support citing sentences.</p><p>A free tool for peer reviewers and authors to verify whether each cited paper supports the sentence citing it. Upload a manuscript; it extracts references, pairs citing sentences, fetches papers, and checks support.</p><p><a href="https://github.com/MarissaFamularo/citation-verifier">MarissaFamularo/citation-verifier</a></p>Evaluation & Benchmarkinghttps://jevbooks.com/en/p/frostney__clean-code-reviewhttps://jevbooks.com/en/p/frostney__clean-code-reviewClean Code review of every PR file, judged by Jev, written by Luna.Sun, 20 Sep 2026 07:16:17 GMT<p>Clean Code review of every PR file, judged by Jev, written by Luna.</p><p>A web tool and MCP server that reviews public GitHub pull requests, diffs or single files against Robert C. Martin's Clean Code. Built for developers who want a checkable, per-file verdict rather than prose alone.</p><p><a href="https://github.com/frostney/clean-code-review">frostney/clean-code-review</a></p>Evaluation & Benchmarkinghttps://jevbooks.com/en/p/anpicasso__hermes-jev-approvalshttps://jevbooks.com/en/p/anpicasso__hermes-jev-approvalsJev-powered smart command approvals for Hermes Agent.Sun, 20 Sep 2026 07:16:16 GMT<p>Jev-powered smart command approvals for Hermes Agent.</p><p>A Hermes Agent plugin that replaces the auxiliary approval reviewer with TypeSafe's Jev decision model, serving only auxiliary.approval when approvals.mode is smart. It returns APPROVE, DENY, or ESCALATE for commands Hermes routes to review.</p><p><a href="https://github.com/anpicasso/hermes-jev-approvals">anpicasso/hermes-jev-approvals</a></p>Verification & Guardrailsflat-fanouttiered-thresholdscode-owned-policyfailure-mode-knobhttps://jevbooks.com/en/p/valentynkit__jev-commithttps://jevbooks.com/en/p/valentynkit__jev-commitPre-commit hook: one Jev call judges commit message vs diff.Sun, 20 Sep 2026 07:16:15 GMT<p>Pre-commit hook: one Jev call judges commit message vs diff.</p><p>A pre-commit hook for developers and agents: it checks whether the commit message matches the staged diff, flags debug leftovers and unmentioned work, and blocks only on a detected credential.</p><p><a href="https://github.com/valentynkit/jev-commit">valentynkit/jev-commit</a></p>Verification & Guardrailsflat-fanouthttps://jevbooks.com/en/p/shiftynick__jev-axihttps://jevbooks.com/en/p/shiftynick__jev-axiAgent-ergonomic CLI for TypeSafe's Jev calibrated judgments.Sun, 20 Sep 2026 07:16:04 GMT<p>Agent-ergonomic CLI for TypeSafe's Jev calibrated judgments.</p><p>jev-axi is a CLI that gives coding agents a second opinion using TypeSafe's Jev, a model that answers typed questions with calibrated probabilities instead of text. It provides commands like pick, rate, check, rank, triage, and guard, plus a safety hook for Claude Code and Codex.</p><p><a href="https://github.com/shiftynick/jev-axi">shiftynick/jev-axi</a></p>Verification & Guardrailsflat-fanouthttps://jevbooks.com/en/p/doeixd__jev-prefhttps://jevbooks.com/en/p/doeixd__jev-prefTurn AGENTS.md preferences into a fast, Jev-powered AI linter.Sun, 20 Sep 2026 07:15:45 GMT<p>Turn AGENTS.md preferences into a fast, Jev-powered AI linter.</p><p>jev-pref is a semantic linter for coding agents. It turns project preferences from AGENTS.md into jev-pref.json rules, checks them against diffs, staged files, or pull requests, and returns findings to the agent.</p><p><a href="https://github.com/doeixd/jev-pref">doeixd/jev-pref</a></p>Verification & Guardrailsrules-as-questionshttps://jevbooks.com/en/p/danrwilloughby__snifftesthttps://jevbooks.com/en/p/danrwilloughby__snifftestProse linter that flags AI writing tells using countable rules and Jev.Sun, 20 Sep 2026 07:15:38 GMT<p>Prose linter that flags AI writing tells using countable rules and Jev.</p><p>Sniff Test reads Markdown and plain text and checks it against house rules, printing one line per flag with file, line, rule and confidence. It is for writers and teams who want to catch AI writing tells.</p><p><a href="https://github.com/DanRWilloughby/snifftest">DanRWilloughby/snifftest</a></p>Evaluation & Benchmarkingscreen-then-deeptiered-thresholdsno-judgment-bandstate-minimalismmeasured-calibrationhttps://jevbooks.com/en/p/vercel-labs__fxhttps://jevbooks.com/en/p/vercel-labs__fxTiny, open, embeddable native coding agent CLI in Zig.Sun, 20 Sep 2026 07:15:37 GMT<p>Tiny, open, embeddable native coding agent CLI in Zig.</p><p>fx is a coding agent CLI written in Zig: a small native binary, open source (Apache-2.0), model-agnostic, and embeddable as a harness in larger systems. Its interface stays closer to a Unix shell than an IDE.</p><p><a href="https://github.com/vercel-labs/fx">vercel-labs/fx</a></p>Verification & Guardrailshttps://jevbooks.com/en/p/coldteadotai__abidehttps://jevbooks.com/en/p/coldteadotai__abideEnforce project rules on every coding agent edit with Jev.Sun, 20 Sep 2026 07:15:28 GMT<p>Enforce project rules on every coding agent edit with Jev.</p><p>Abide is a TypeScript CLI that supervises coding agents (Claude Code, Codex, OpenCode). It reads every edit and turn, checks them against rules from your AGENTS.md/CLAUDE.md, and makes the agent repair violations in the same turn.</p><p><a href="https://github.com/coldteadotai/abide">coldteadotai/abide</a></p>Verification & Guardrailsstate-minimalismhttps://jevbooks.com/en/p/kelbie__hunchhttps://jevbooks.com/en/p/kelbie__hunchSemantic code review with Jev, plain-English rules and Agent Skills.Sun, 20 Sep 2026 07:15:26 GMT<p>Semantic code review with Jev, plain-English rules and Agent Skills.</p><p>Hunch lets a person or coding agent ask plain-English questions across a repository and get scored source locations to investigate. It also runs recurring review rules on diffs, locally or on pull requests.</p><p><a href="https://github.com/Kelbie/hunch">Kelbie/hunch</a></p>Scoring & Rankinghttps://jevbooks.com/en/p/akashpriyadarshii__jev-githttps://jevbooks.com/en/p/akashpriyadarshii__jev-gitSub-second Git hook screening staged diffs with Jev.Sun, 20 Sep 2026 07:15:21 GMT<p>Sub-second Git hook screening staged diffs with Jev.</p><p>jev-git is a Git pre-commit and pre-push hook for developers. It screens staged diffs for unredacted secrets, prompt-injection attacks, and destructive commands before they enter git history, running in ~80ms as a single static Rust binary.</p><p><a href="https://github.com/AkashPriyadarshii/jev-git">AkashPriyadarshii/jev-git</a></p>Verification & Guardrailshttps://jevbooks.com/en/p/devagrawal09__stanley-codehttps://jevbooks.com/en/p/devagrawal09__stanley-codeBounded TypeSafe Jev workflows for coding agents.Sun, 20 Sep 2026 07:15:16 GMT<p>Bounded TypeSafe Jev workflows for coding agents.</p><p>Stanley is a command-line tool for checking and understanding code changes. You describe what you need in natural language; it routes the request to one bounded workflow that gathers evidence and asks Jev fixed-choice questions.</p><p><a href="https://github.com/devagrawal09/stanley-code">devagrawal09/stanley-code</a></p>Classification & Routinghttps://jevbooks.com/en/p/thruwire__foremanhttps://jevbooks.com/en/p/thruwire__foremanSoftware factory foreman using Jev to supervise coding workers.Sun, 20 Sep 2026 07:15:07 GMT<p>Software factory foreman using Jev to supervise coding workers.</p><p>Foreman is a Python asyncio runtime that sits above Codex or OpenCode workers. It takes a free-form software job and independently assesses whether the implementation is complete, tests are sufficient, verification is needed, or a human is required.</p><p><a href="https://github.com/thruwire/foreman">thruwire/foreman</a></p>Verification & Guardrailsflat-fanoutcross-checkhttps://jevbooks.com/en/p/leepokai__jev-guardhttps://jevbooks.com/en/p/leepokai__jev-guardAuto mode security guard for coding agents, powered by Jev.Sun, 20 Sep 2026 07:15:04 GMT<p>Auto mode security guard for coding agents, powered by Jev.</p><p>jev-guard is a security hook for coding agents including Claude Code, Codex, Copilot, Gemini, Cursor, pi, OpenCode, and ACP clients. It risk-scores tool calls, flags prompt injection in results, and checks skills and plugins.</p><p><a href="https://github.com/leepokai/jev-guard">leepokai/jev-guard</a></p>Verification & Guardrailshttps://jevbooks.com/en/p/y0usaf__pi-jevhttps://jevbooks.com/en/p/y0usaf__pi-jevTypeSafe Jev as a decision layer for the Pi coding agent.Sun, 20 Sep 2026 07:14:59 GMT<p>TypeSafe Jev as a decision layer for the Pi coding agent.</p><p>Adds a measured tool-call gate to the Pi coding agent so risky calls are checked by Jev before execution. Also judges bash output and provides a jev_ask tool for typed, calibrated answers.</p><p><a href="https://github.com/y0usaf/pi-jev">y0usaf/pi-jev</a></p>Verification & Guardrailsflat-fanouttiered-thresholdsfailure-mode-knobhttps://jevbooks.com/en/p/devagrawal09__jev-reviewhttps://jevbooks.com/en/p/devagrawal09__jev-reviewStaged code-review workflow with a local dashboard, gated by Jev.Sun, 20 Sep 2026 07:14:58 GMT<p>Staged code-review workflow with a local dashboard, gated by Jev.</p><p>A small TypeScript code-review workflow for developers: it reviews a Git diff or scans a whole codebase and shows results in a quiet local dashboard bound to 127.0.0.1.</p><p><a href="https://github.com/devagrawal09/jev-review">devagrawal09/jev-review</a></p>Evaluation & Benchmarkinghttps://jevbooks.com/en/p/luantak__is-malicioushttps://jevbooks.com/en/p/luantak__is-maliciousScan a codebase for malicious behavior with TypeSafe Jev.Sun, 20 Sep 2026 07:14:50 GMT<p>Scan a codebase for malicious behavior with TypeSafe Jev.</p><p>A CLI and GitHub Action that scans source, configuration, build, and CI files for hidden, deceptive, or data-stealing behavior, giving a second opinion before running unfamiliar code.</p><p><a href="https://github.com/luantak/is-malicious">luantak/is-malicious</a></p>Verification & Guardrailshttps://jevbooks.com/en/p/bohutang__sifthttps://jevbooks.com/en/p/bohutang__siftChrome extension labels X posts with Jev and hides unwanted ones.Sun, 20 Sep 2026 07:14:47 GMT<p>Chrome extension labels X posts with Jev and hides unwanted ones.</p><p>Sift is a Chrome extension that labels every post and reply on X with one of six categories (Substance, Humor, Chit-chat, Promo, Junk, AI-written) plus two flags (AI-written, Off-topic). Users choose which labels to hide, making the feed tech-only by default.</p><p><a href="https://github.com/bohutang/sift">bohutang/sift</a></p>Data Labeling & Curationhttps://jevbooks.com/en/p/matthewp__flue-jev-demohttps://jevbooks.com/en/p/matthewp__flue-jev-demoFlue agent routing with TypeSafe Jev via Cloudflare AI GatewaySun, 20 Sep 2026 07:14:42 GMT<p>Flue agent routing with TypeSafe Jev via Cloudflare AI Gateway</p><p>A standalone Cloudflare-targeted Flue example in TypeScript that routes a support agent's work. It is for developers exploring Flue agents on Cloudflare Workers who want typed routing decisions from Jev alongside a chat model.</p><p><a href="https://github.com/matthewp/flue-jev-demo">matthewp/flue-jev-demo</a></p>Classification & Routinghttps://jevbooks.com/en/p/nexibeo__jev-cookbookhttps://jevbooks.com/en/p/nexibeo__jev-cookbook15 runnable Node recipes for Jev decision model on OpenRouter.Sun, 20 Sep 2026 07:14:40 GMT<p>15 runnable Node recipes for Jev decision model on OpenRouter.</p><p>A cookbook of 15 practical, tested recipes for TypeSafe's Jev decision model. Each recipe includes a runnable script, a small labelled dataset, and measured results for tasks like support triage, file organizing, and Gmail labeling.</p><p><a href="https://github.com/nexibeo/jev-cookbook">nexibeo/jev-cookbook</a></p>Classification & Routingflat-fanouthttps://jevbooks.com/en/p/islee23520__omo-jevlike-routerhttps://jevbooks.com/en/p/islee23520__omo-jevlike-routerOne-pass skill router that shrinks OmO's system-prompt skill catalog.Sun, 20 Sep 2026 07:14:39 GMT<p>One-pass skill router that shrinks OmO's system-prompt skill catalog.</p><p>A Jev-style skill router for OmO. A frozen Qwen2.5-0.5B plus a small option-attention head scores the whole skill catalog against the user's prompt, and an OmO extension trims the skill block to the top-K.</p><p><a href="https://github.com/islee23520/omo-jevlike-router">islee23520/omo-jevlike-router</a></p>Classification & Routingscreen-then-deepfailure-mode-knobmeasured-calibrationhttps://jevbooks.com/en/p/ffattiger__new-api-plugin-typesafehttps://jevbooks.com/en/p/ffattiger__new-api-plugin-typesafeTypeSafe AI System One task plugin for new-api gateway.Sun, 20 Sep 2026 07:14:34 GMT<p>TypeSafe AI System One task plugin for new-api gateway.</p><p>A new-api task plugin that adds a native /v1/systemone endpoint, letting clients send typed questions to Jev through the same gateway used for chat models.</p><p><a href="https://github.com/FFatTiger/new-api-plugin-typesafe">FFatTiger/new-api-plugin-typesafe</a></p>Evaluation & Benchmarkingflat-fanouthttps://jevbooks.com/en/p/abovecolin__ha-jevhttps://jevbooks.com/en/p/abovecolin__ha-jevAsk your house a question, get a number back.Sun, 20 Sep 2026 07:14:26 GMT<p>Ask your house a question, get a number back.</p><p>A Home Assistant integration that turns TypeSafe Jev questions about your home into sensors and automation actions, for people who want typed, probabilistic answers rather than chat.</p><p><a href="https://github.com/AboveColin/HA-Jev">AboveColin/HA-Jev</a></p>Infra / SDKs / Integrationshttps://jevbooks.com/en/p/raihankhan-rk__diffjuryhttps://jevbooks.com/en/p/raihankhan-rk__diffjuryJev routes PR risk before human review.Sun, 20 Sep 2026 07:14:20 GMT<p>Jev routes PR risk before human review.</p><p>DiffJury is a web app that fetches a public GitHub pull request and uses Jev to judge merge risk, review depth, and whether design or security review is needed. It is for teams that want to triage PRs before assigning a human reviewer.</p><p><a href="https://github.com/raihankhan-rk/diffjury">raihankhan-rk/diffjury</a></p>Classification & Routingflat-fanouthttps://jevbooks.com/en/p/realzachi__typesafe-adblockhttps://jevbooks.com/en/p/realzachi__typesafe-adblockChrome extension asks Jev if each DOM element is an adSun, 20 Sep 2026 07:14:13 GMT<p>Chrome extension asks Jev if each DOM element is an ad</p><p>A fun Chrome extension (Manifest V3) that detects ad-shaped DOM elements in real time and removes them. Intended for people curious about using a small decision model, not as a real ad blocker.</p><p><a href="https://github.com/realZachi/typesafe-adblock">realZachi/typesafe-adblock</a></p>Classification & Routingflat-fanouthttps://jevbooks.com/en/p/kitze__unclutterhttps://jevbooks.com/en/p/kitze__unclutterWXT browser extension that removes page clutter using Jev.Sun, 20 Sep 2026 07:14:05 GMT<p>WXT browser extension that removes page clutter using Jev.</p><p>A Chrome/Chromium and Firefox extension that classifies nonessential page elements as clutter and hides them, storing reusable hiding rules per page template for the user.</p><p><a href="https://github.com/kitze/unclutter">kitze/unclutter</a></p>Agent Decisionshttps://jevbooks.com/en/p/giesn__typesafe-jev-workflowhttps://jevbooks.com/en/p/giesn__typesafe-jev-workflowAsync LangGraph workflow routes emails via Jev Choice.Sun, 20 Sep 2026 07:14:01 GMT<p>Async LangGraph workflow routes emails via Jev Choice.</p><p>A small async LangGraph workflow that sends a mocked email to TypeSafe's Jev model, receives a typed Choice (invoice or general), and routes to a demo handler. Handlers only set a destination in graph state; they do not send email or make payments.</p><p><a href="https://github.com/GiesN/typesafe-jev-workflow">GiesN/typesafe-jev-workflow</a></p>Classification & Routingfailure-mode-knobhttps://jevbooks.com/en/p/gtaras7__typesafe-jevhttps://jevbooks.com/en/p/gtaras7__typesafe-jevScreen a folder of CVs with typed Jev judgments.Sun, 20 Sep 2026 07:13:51 GMT<p>Screen a folder of CVs with typed Jev judgments.</p><p>A local screening workbench for a folder of CVs, built on TypeSafe's Jev model. It is for anyone hiring who wants a sortable shortlist with one verdict per candidate.</p><p><a href="https://github.com/gtaras7/typesafe-jev">gtaras7/typesafe-jev</a></p>Scoring & Rankingcode-owned-policyhttps://jevbooks.com/en/p/godsboy__jev-agent-skill-routerhttps://jevbooks.com/en/p/godsboy__jev-agent-skill-routerTyped, confidence-aware agent skill routing with TypeSafe Jev.Sun, 20 Sep 2026 07:13:49 GMT<p>Typed, confidence-aware agent skill routing with TypeSafe Jev.</p><p>A Python CLI and library that routes a request to one skill from a catalogue of skill names and public descriptions, returning route, no_skill, or review. It is an optional routing layer for agent platforms like Hermes Agent.</p><p><a href="https://github.com/GodsBoy/jev-agent-skill-router">GodsBoy/jev-agent-skill-router</a></p>Classification & Routinghttps://jevbooks.com/en/p/adarshmishra07__jcm-routerhttps://jevbooks.com/en/p/adarshmishra07__jcm-routerLocal proxy picks Claude model and effort per message with Jev.Sun, 20 Sep 2026 07:13:48 GMT<p>Local proxy picks Claude model and effort per message with Jev.</p><p>jcm-router is a local proxy between Claude Code and the Anthropic API for developers using a Claude subscription. It routes subagents and, when cheap, the main chat to different Claude models and effort levels.</p><p><a href="https://github.com/adarshmishra07/jcm-router">adarshmishra07/jcm-router</a></p>Classification & Routinghttps://jevbooks.com/en/p/usenotra__notrahttps://jevbooks.com/en/p/usenotra__notraTrack brand visibility in AI answers and draft content.Sun, 20 Sep 2026 07:13:40 GMT<p>Track brand visibility in AI answers and draft content.</p><p>Notra is a generative engine optimization (GEO) platform that runs buyer questions across AI engines, tracks mentions, citations, and competitors, and turns content gaps into drafts.</p><p><a href="https://github.com/usenotra/notra">usenotra/notra</a></p>Classification & Routinghttps://jevbooks.com/en/p/gargpratyush__jev-routerhttps://jevbooks.com/en/p/gargpratyush__jev-routerRoute Claude Code and Codex turns to the cheapest capable model via Jev.Sun, 20 Sep 2026 07:13:39 GMT<p>Route Claude Code and Codex turns to the cheapest capable model via Jev.</p><p>jev-router is a developer tool that adds automatic per-turn model routing to Claude Code and OpenAI Codex. It launches the real upstream CLI and uses Jev to send simple work to a fast tier and difficult work to a strong tier.</p><p><a href="https://github.com/gargpratyush/jev-router">gargpratyush/jev-router</a></p>Classification & Routinghttps://jevbooks.com/en/p/prismhq__jev-routerhttps://jevbooks.com/en/p/prismhq__jev-routerOpen-source LLM router where Jev picks the serving model.Sun, 20 Sep 2026 07:13:39 GMT<p>Open-source LLM router where Jev picks the serving model.</p><p>An OpenAI-compatible LLM router built on LiteLLM. Clients send one model id, &quot;jev-router&quot;, and the router selects which underlying model serves each request. For developers who want automatic model selection across providers like OpenRouter.</p><p><a href="https://github.com/prismhq/jev-router">prismhq/jev-router</a></p>Classification & Routinghttps://jevbooks.com/en/p/mejiasd3v__pi-jev-routerhttps://jevbooks.com/en/p/mejiasd3v__pi-jev-routerJev picks a Pi model and effort, then pins them.Sun, 20 Sep 2026 07:13:39 GMT<p>Jev picks a Pi model and effort, then pins them.</p><p>A Pi coding agent extension that adds automatic per-request model routing. Jev chooses a model and reasoning effort once per session, then both stay fixed while generation uses your existing Pi providers and credentials.</p><p><a href="https://github.com/mejiasd3v/pi-jev-router">mejiasd3v/pi-jev-router</a></p>Classification & Routing