jev.directory

Systems

16 systems shared by the community.

Qualitative query

suffername 30 downloads Find source passages with saved questions and Jev scoring, then connect them with native block embeds. Ask your vault a question. Conne…

eslint.config.mjs
import { defineConfig } from 'eslint/config';
import obsidianmd from 'eslint-plugin-obsidianmd';

export default defineConfig([
  { ignores: ['dist/**', 'release/**', 'node_modules/**', 'scripts/**', 'tests/**', '*.mjs'] },
  ...obsidianmd.configs.recommended,
  {
    languageOptions: { parserOptions: { projectService: true } },
    rules: {
      "obsidianmd/ui/sentence-case": ["warn", {
        brands: ["Jev", "TypeSafe", "Qualitative Query", "Obsidian"],
        // Preserve the credential prefix and the established UI spelling for block IDs.

Filter advertising information using the JEV model.

JevNoiseGate filters unwanted notifications and SMS on Android. Rather than matching keywords, an LLM decides what's noise — and only what it explicitly flags…

app/src/main/kotlin/me/ethanxu/jevnoisegate/app/di/BackendModule.kt
package me.ethanxu.jevnoisegate.app.di

import android.os.Build
import dagger.Module
import dagger.Provides
import dagger.hilt.InstallIn
import dagger.hilt.components.SingletonComponent
import javax.inject.Singleton
import me.ethanxu.jevnoisegate.app.BuildConfig
import me.ethanxu.jevnoisegate.core.decision.DecisionBackend
import me.ethanxu.jevnoisegate.core.decision.TypeSafeBackend
import me.ethanxu.jevnoisegate.core.data.settings.SettingsRepository
Uufec
noulscorechoiceClassificationTriageVerification

RedReplier buying signal ranker

Rank intent and product fit, flag when a competitor gets named, weigh the poster's reach, and tell you which threads to reply to now and which to just watch. S…

JEV is INSANE 🤯 We gave it 1.7 million mentions In 53 seconds, it scored 1,759,932 buying signals across Reddit, X, Bluesky, Hacker News and Facebook, and drafted 6,752 replies in the voice of your last 20 posts. All for just $0.65. JEV can also rank intent and product fit,

RouteKit — Jev reads the task, a policy engine picks the model

Model routing for agents, with the judgment and the decision deliberately kept apart. The premise is that asking a full LLM which LLM should handle a request i…

src/decision/jev/jev-normalizer.ts
import { RoutingSignals } from '../../types/signals';
import { ResultFor } from '@typesafe-ai/sdk';
import { routingQuestions } from './jev-questions';

type JevResponses = {
  [K in keyof typeof routingQuestions]: ResultFor<typeof routingQuestions[K]>;
};

/**
 * Normalizes a 0-4 scale score to a 0.0-1.0 range.
 */
function normalizeScore(rawScore: number): number {

TypeSafe Mario — Jev picks NES controller inputs from telemetry, not pixels

An experimental controller that lets Jev choose the next NES input for the original Super Mario Bros. The interesting constraint is what the model is not given…

src/typesafe_mario/__init__.py
"""TypeSafe-powered direct controller for Super Mario Bros."""

from .actions import Action
from .state import MarioSnapshot, MarioStateParser

__all__ = ["Action", "MarioSnapshot", "MarioStateParser"]

Jev Review — a staged code review that narrows before it scores

A code-review workflow that reviews a Git diff or scans a whole codebase, built as a pipeline of bounded Jev judgments rather than one big ask. The stages run…

src/dashboard/server.ts
// Local-only dashboard server: a fixed list of static assets from ./public
// plus the saved review report as JSON. Binds to loopback only.
import { readFile } from "node:fs/promises";
import { createServer, type IncomingMessage, type ServerResponse } from "node:http";
import { join, relative } from "node:path";
import { readReport, reportPath } from "../adapters/report-store.ts";

const HOST = "127.0.0.1";
const PORT = Number(process.env.PORT ?? 4317);
const REPORT = reportPath();
const PUBLIC_DIR = join(import.meta.dirname, "public");

Foreman — a fast judgment model supervising slow coding agents

Foreman puts Jev above the coding agents rather than inside them. Codex workers do the software engineering; Foreman watches the factory floor and independentl…

src/foreman/cli.py
from __future__ import annotations

import asyncio
import os
from datetime import UTC
from pathlib import Path
from typing import Annotated

import typer
from dotenv import load_dotenv
from rich.console import Console
from rich.table import Table

Foreman — a fast judgment model supervising slow coding agents

Foreman puts Jev above the coding agents rather than inside them. Codex workers do the software engineering; Foreman watches the factory floor and independentl…

src/foreman/cli.py
from __future__ import annotations

import asyncio
import os
from datetime import UTC
from pathlib import Path
from typing import Annotated

import typer
from dotenv import load_dotenv
from rich.console import Console
from rich.table import Table

Jev Ultrafast — a browser agent whose action space is rebuilt every step

A browser agent from the Browser Use team. Every observation produces a fresh element table, and Jev picks an operation and a target over that table — not over…

docs/flights-measurement.json
{
  "elapsed_ms": 7073,
  "decision_requests": 17,
  "decision_median_ms": 178,
  "decision_total_ms": 3720,
  "browser_actions": 11,
  "wait_actions": 1,
  "models": {
    "decision": [
      "jev-1.13.0"
    ],
    "text": [

Jev Browser — headless browser use, one Jev judgment per step

jev-browser drives a real headless Chromium and asks Jev to pick the next action, one call per step. Three judgments over the same state, deliberately unable t…

scripts/ensure-chromium.mjs
// Downloads Chromium for Playwright unless it is already present or the
// caller opted out. Keeps `npx -y github:jkudish/jev-browser` self-contained:
// "packages everything it needs to navigate directly".
import { execFileSync } from "node:child_process";

if (process.env.JEV_BROWSER_SKIP_BROWSER_DOWNLOAD === "1" || process.env.PLAYWRIGHT_SKIP_BROWSER_DOWNLOAD === "1") {
  process.exit(0);
}

try {
  execFileSync("npx", ["--yes", "playwright", "install", "chromium"], {
    stdio: "inherit",

I gave Jev 3,282 of my X posts across 100M views and asked it to find what ac…

Ian Nuttall ran 3,282 of his own X posts (100M views) through Jev to find what actually drives growth. Each post was asked eight questions — its topic, its hoo…

I gave Jev 3,282 of my X posts across 100M views and asked it to find what actually works for growth. 4,252,330 tokens $0.1282 for the full 8m 34s run! Each post got 8 questions about the topic, hook, tone, whether it teaches something, etc. How-to posts got 150 median likes

Celesto

Jev helps speed up PR review with rubric based eval

examples/pr-review-jev/config.py
"""Resolve provider credentials without silently switching providers during a review."""

import os


def llm_config(env=None):
    env = os.environ if env is None else env
    provider = env.get("LLM_PROVIDER") or (
        "custom" if env.get("LLM_BASE_URL") else "openai" if env.get("OPENAI_API_KEY") else "gateway"
    )
    if provider == "openai":
        base, key, default = "https://api.openai.com/v1", env.get("OPENAI_API_KEY"), "gpt-5.6-luna"

I made a DuckDB extension where you can use Jev to do quick classification

Hamilton Ulmer created a DuckDB quick classification where you can use @typesafeai's Jev to do quick classification of rows in any csv/parquet file or duckdb t…

I made a DuckDB extension where you can use @typesafeai 's Jev to do quick classification of rows in any csv/parquet file or duckdb table about 10sec for 1k rows ~ better than using an LLM, way more ergonomic than a classifier game-changing for data analysis!

Jev shadcn lint eval

This checks shadcn-ui/lint (commit 53de86f) with TypeSafe's Jev model (jev-latest). The repo's own eval measures how agents style UI before and after lint feed…

build-states.mjs
#!/usr/bin/env node
// Turns a shadcn-ui/lint eval run into System One states.
//
// Usage: node build-states.mjs <results.json> [--out states.jsonl]
//
// One "finding" state per condition-A lint finding (the diagnostics the
// agent was fed) and one "task" state per task (A vs C sources). The
// project's tokens and components come from design-system.json, keyed by
// the run's fixture (packages/evals/<fixture>).
import * as fs from "node:fs"
import * as path from "node:path"
import { fileURLToPath } from "node:url"
Bblas0
noulscorechoice