A contextual brand-suitability scorer that judges each feed item against an advertiser's own versioned policy, returning a category, an exclusion flag, and a confidence for routing borderline items to review.
A pre-bid spend guard that scores a placement's expected relevance and flags whether it matches the campaign brief, letting deterministic code hold budget back before the auction instead of gating the creative itself.
A real-estate matching tool that applies hard-constraint gates before ranking a buyer's property preferences against listing features and remarks, returning a fit band and a per-feature score for the agent.
A travel rebooking demo that classifies a trip disruption, checks itinerary feasibility, and ranks concrete rebooking options against a traveler's ordered preferences, returning ordered choices with probabilities.
A playable espresso bar where an XState machine models what the hardware can do next and Jev picks the move, from a plain-language order or after you break the machine.
A Claude Code plugin that replaces the built-in auto-mode permission classifier with Jev, deciding tool calls in about half the time while agreeing with the original on the calls it handles.
Cloudflare Worker that asks Jev on Workers AI for literal profanity in text or usernames and, for usernames, a second Noul for phonetic or look-alike disguises. The threshold and max policy live in Worker code, with OpenAPI docs generated via chanfana and Hono.
Third-party TypeScript check that compares Jev (typesafe-ai/jev) against openai/gpt-4o-mini and anthropic/claude-sonnet-4.5 under identical conditions, routing 60 synthetic booking inquiries in four languages for a photo-shoot service, all through one Vercel AI Gateway key.
Terminal agent for Polymarket's BTC Up or Down five-minute markets. Roughly every five seconds it asks Jev whether BTC finishes UP or DOWN versus the window open, then abstains or buys once and holds to resolution. Defaults to dry-run; live orders are opt-in.
Real-time geopolitical intelligence dashboard that uses TypeSafe Jev questions to score news headline severity into five threat levels and categorize events across 14 conflict, cyber and infrastructure domains, alongside news aggregation and a Country Instability Index.
Open-source link attribution platform that calls typesafe-ai/jev in malicious-link-check.ts before a short link is created, so the URL is gated by a typed Jev verdict rather than a static blocklist. Next.js and TypeScript with an open-core license.
A bring-your-own-key Chrome extension that asks Jev one slop/not_slop Choice per X or LinkedIn post as it scrolls into view, then blurs and stamps anything at or above a configurable threshold. Scrolling never blocks, and each post is scored once from a local cache.
An MIT-licensed, local-first conversation memory app that transcribes meetings, calls and voice memos on-device into Markdown you own. Its live voice path runs evaluations through Jev, and 34 MCP tools plus a 58-command CLI let coding agents query the history you choose to share.
A browser extension plus Cloudflare Worker that scores search results before you click: Jev answers six questions per page (relevance, depth, SEO filler, sales page, page type, key passage) and code combines them into a 0-100 badge that re-ranks Google, Bing, DuckDuckGo and 104.com.tw.
A DeepSeek Harness plugin where Jev Choice and Noul checks gate every non-read-only tool call - deny confident high-risk calls, escalate ambiguous ones, auto-approve clearly granted reversible ones - and optionally score session messages so valuable context survives instead of being dropped oldest-first.
OpenClaw plugin and CLI that turn a plain-language --when/--not-when condition into a scheduled trigger script. Each tick sends one Noul through OpenClaw's decisionModel and only wakes the conversation model when the condition reaches 0.7 or above; a 76-tick benchmark reports 75 correct judgments and 0 false wake-ups.
MCP server and Claude Code plugin that files each captured note, link or meeting line with one Jev request: kind, project, action and duplicate questions answered in parallel. Routes to a project only at 0.45 or above, marks a duplicate only at 0.70 with a matching item, and stores text verbatim in append-only JSONL.
Verifiable coding agent runtime with 855 stars. Ships a first-class Jev structured-output provider so agent decisions come back typed through the same decision resolver as its other providers, alongside plain-English workflows, approval gates and evidence-backed verification stages.
Open-source, pre-alpha CLI that runs plain-English browser tests on Playwright, using Jev for bounded decisions: resolving each step, picking elements among offered candidates and judging verify claims with probabilities. Bring your own TypeSafe key; MIT licensed.
System One Mosaic Benchmark: a Python evaluator and Next.js viewer comparing Jev and open decision models across 137 English Choice, Noul and Score benchmarks from 106 dataset subsets, with recorded evaluation conditions and a public Hugging Face leaderboard.
Interactive media artwork where one post meets 100 fictional observers: for each observer the app asks Jev one Choice plus four Score questions (500 questions total), animates the eyes in Three.js and later exposes the full distribution of reactions in a post-analytics view.
Cline's official curated plugin collection for its CLI and extensions. Includes jev-browser, which executes browser goals with Jev through Vercel AI Gateway plus screenshots and isolated Chromium sessions; plugins install by directory slug, e.g. cline plugin install jev-browser.
Open-source workspace with 29,796 stars for building, deploying and monitoring AI agents and workflows -- visual builder, chat, tables, files and knowledge -- featuring native TypeSafe System One evaluation and decision provider integration.
x402 risk-check provider that scores agent counterparties before payment: OFAC, phishing and drainer feeds, transaction simulation, drainer-kit fingerprints and a Jev model reading context for injected instructions, issuing an ES256-signed attestation per verdict.
pi coding-agent package that gates subagent dispatches: the task text is evaluated in one typed Jev call against configurable rules (per-rule probability questions with below/above thresholds), violations block the dispatch with per-rule reasons, and errors fail open to allow.
MCP server that lets Claude Code, Cursor and other clients drive Chrome over the DevTools Protocol, exposing an accessibility-tree menu of stable refs. Its example Jev loop has Jev pick browser actions from that menu, and the README reports about a third fewer tokens and a quarter fewer tool calls in a blind benchmark.
Open-source TypeScript agentic operating system: a core runtime, app for web and mobile, CLI, cloud services and first-party plugins. The directory's curated summary reports it integrates TypeSafe System One decision services for sub-100ms intent classification, action dispatching and confidence-gated tool execution.
Claude Code plugin that compacts conversations by moving old tool results to local files instead of summarizing them, leaving a ticket line in their place. When the agent needs something back, Jev can choose among the moved-out items in words for a `find`, while `recall` restores any item exactly by id.
React dashboard and simulator for a 10-million smart-meter fleet that triages telemetry through a deterministic edge pre-filter and then Jev Choice, Score and probability primitives, triggering SIM quarantine, work orders or RF fixes at about $33 per month in its own model.
Fuzzy linter hook for Claude Code and Codex: after the agent edits a file, it sends the added code to Jev, which answers Noul checks against team rule packs in about 0.3 seconds, letting the agent fix violations before code review.
A Pi coding-agent extension that uses Jev to classify each request and route it to one of three configured models. Supports TypeSafe, Cloudflare AI Gateway, Vercel AI Gateway and OpenRouter backends, with doctor checks, shadow mode and an off-by-default setup.
A teams-first multi-agent orchestration plugin for Claude Code with opt-in Jev hooks for sub-millisecond judgment points, decision caching and per-point egress controls. TypeScript, MIT, with a staged team pipeline and tmux CLI workers.
A local Jev-compatible server that reads Choice, Score and Noul answers from the logits of any MLX checkpoint or OpenAI-compatible endpoint with no training, serving POST /v1/systemone as a drop-in for the official TypeSafe SDK.
A CLI, Claude Code plugin and Pi extension that pick a model per task from one Jev call: a Choice over candidates asked in both orders plus a complexity Score that sets a minimum tier, with quota-aware filtering and fallback.
A Next.js dashboard and daily Vercel cron that scores public job postings against one resume with Jev through the Vercel AI Gateway, then applies through a Grok Bot webhook that fills ATS forms and writes status back.
Chrome MV3 extension plus a local MCP server exposing 50 browser tools; its chrome_act_toward_goal micro-loop uses TypeSafe Jev System One or a local decider to pick the next action on an 85%+ pruned, Shadow-DOM-pierced DOM tree, dispatching trusted CDP events in logged-in Chrome without stealing focus.
CLI-installed Gmail classifier that runs 24/7 on user-owned Google Apps Script: the worker asks Jev for three typed judgments per thread (requires_action, is_important, bucket), applies labels, stars, and optional archiving, and routes ambiguous or low-confidence mail to a Review label.
Open-source AI email assistant (Next.js, Postgres, 12k+ stars) with a decision-model layer that can run on openrouter:typesafe/jev-1.13: Jev-backed questions cover rule selection, cold-email detection, thread status, recurring patterns, unsubscribe pages, and sender categorization, each with evals.
Merged PR in the Kilo Code cloud monorepo that replaces its LLM auto-routing classifier with one OpenRouter /systemone Choice request to typesafe/jev-1.13; the PR reports 0.93-0.94 route accuracy on 72 golden cases at roughly $0.00028 per call, with a 250 ms transport-error retry.
Reproducible benchmark asking which approved agent answers changed meaning after a config edit: one Jev Choice (same, fact_differs, action_differs, specificity_differs) caught all 19 real changes with 13 false alarms across 109 before/after pairs, versus 33 for a three-tier stack and 19 for an LLM judge.
Three.js simulation where an OpenAI model turns each plain-language instruction into a plan and Jev makes every reflex decision: two Score questions for move and turn plus one Noul for jump, with code acting only past a 0.33 dead zone and sending no Jev request while no step is active.
Permission gate extension for the pi coding agent: deterministic allow/deny rules and protected paths settle clear cases first, then gray-zone tool calls get one structured classify() verdict; Jev is an optional native classifier via TypeSafe direct, OpenRouter, or other gateways, failing closed on error.
OpenCode plugin that replaces LLM compaction summaries with Jev keep/drop judgments: stale tool calls and outputs are pruned while user and assistant text is preserved verbatim, falling back to the built-in summary when Jev cannot help.
Novelty npm package whose crypto.randomUUID() look-alike sends 15 Noul questions plus one Choice to Jev and packs the returned probabilities into the bytes of a UUID v4-shaped string; explicitly not cryptographically random.
Privacy-first, self-hosted personal knowledge management system with block-level references and a local-first Markdown editor; the project integrates Jev's decision model for high-speed document classification, flashcard intent categorization and automated tag routing.
Dependency-free TypeScript SDK that turns Jev Noul, Choice and Score answers into named decisions: enter/exit thresholds with hysteresis, nested decision trees settled in one call, a JSONL journal, threshold replay without inference and Brier calibration.
Claude Code plugin and CLI that runs Claude Code on Opus 5.5 and has Jev re-decide the reasoning effort when a prompt arrives and after every tool batch, sending each change as a per-message effort statement inside one unbroken prompt cache.
Fixed-wing landing simulator where a human can compete against Jev: four Jev Choice decisions map to aileron, elevator, rudder and throttle commands in JSBSim, run in real time. The benchmark also compares Jev, classical controllers and Codex models on landing score and API cost.