jev-browse

Where it calls Jev

import { TypeSafeClient } from "@typesafe-ai/sdk";

jev-browse.mjs:6

The link points at the commit we read, so the line number still holds.

What it asks Jev

These question sets are lifted from this project's source exactly as written.

import { experimental_evaluate as evaluate } from 'ai';

const { answers } = await evaluate({
  model: 'typesafe-ai/jev',
  state,
  questions: {
    done: { type: 'boolean', instructions: 'Is `current_page` the page that `goal` asks to reach?' },
    blocked: { type: 'boolean', instructions: 'Does reaching `goal` from `current_page` require an action listed in `agent.cannot`?' },
  },
});

Other projects in this category

• jev-playground ▶ — TypeSafe AI の System One モデル Jev を MoonBit から触るためのプレイグラウンド。
• jev-as-a-judge ▶ — Using Jev as an evaluator.
• eutrya ▶ — Jev-native AI security harness for autonomous research, multi-agent swarms, persistent hunt boards, and long-running agent workflows. CLI-first, open source, and built for authorized security research.
• typesafe-local ▶ — Inspired by TypeSafe Ai, Ask a local LLM typed questions, get calibrated probabilities instead of text. Structured output without generation or parsing. MLX / Apple Silicon.
• jev-mcp ▶ — Connect JEV to MCP clients and compare its judgments against general-purpose LLMs using shared datasets and measurable accuracy.
• omp-jev ▶ — TypeSafe Jev routing for Oh My Pi, with an opt-in checkpoint orchestrator and editable XDG configuration. Requires Bun = 1.3.14 and OMP = 18.2.3.
• jev-chat ▶ — A chatbot from typed Jev decisions: hierarchical speculative decoding over System One probabilities.
• jev-broadcast-lab ▶ — Operator lab for TypeSafe Jev. Chess Arena, closed-schema booths, Stockfish HUD for review only.