Skip to content

typesafe-arena

Local markdown mirror of TypeSafe's docs with a Rust BM25 and knowledge-graph search tool, Lume.

Appprototype

This repository is mostly documentation or write-up rather than runnable code.

Overview

Although listed as a Jev playground, the repository's README describes a local markdown mirror of docs.typesafe.ai, crawled on 17 September 2026, plus Lume, a Rust search tool. Lume builds a BM25 index and a semantic knowledge graph from a tag dictionary of TypeSafe terms such as intent routing, Choice, Noul and confidence, so searches walk related concepts.

The mirrored docs cover patterns (composite scoring, confidence-gated routing, speculative fan-out, intent routing), primitives and concepts. It is useful for offline reading or for giving a local agent searchable TypeSafe documentation, but it does not itself call Jev. Lume can optionally add dense vectors or LLM entity extraction via Ollama, but lexical plus graph search works without them. The README reports that a search for intent routing ranked the matching pattern page first.

What the decision model decides

TypeQuestionAnswers
OtherNone; it indexes documentation about Choice, Score and Noul.(inferred)—

How it works

  1. Build Lume with Cargo.
  2. Index the mirrored docs with the TypeSafe tag dictionary.
  3. Search with lexical plus graph boost.

Strengths

  • Offline, searchable copy of TypeSafe docs.
  • Graph search links related concepts.

Caveats

  • A snapshot that may drift from current docs.
  • Does not call Jev or run decisions.

Getting started

cd lume && cargo build --release, then lume index --db .lume-index --tag-dict dict/typesafe.csv docs

At a glance

Type
App
Status
prototype
Language
Rust
Runtime
CLI
Has code
yes

Use cases and integrations

How curated lists describe it

Frequently asked questions

Is this a Jev playground?

The README shows a docs mirror with a search tool rather than a playground.

How current are the docs?

They were crawled on 17 September 2026.

What is Lume?

A Rust BM25 plus knowledge-graph search tool bundled in the repository.

Related projects

  • Claude Code plugin that finds fuzzy decision points in an app and ranks where a Jev question could replace them, with precedents.

    ProjectbetaPython
  • Agent skill for writing and improving programs that call Jev: question design, state structure, thresholds and diagnosing wrong answers.

    Agent skillproduction
  • Server behind api.typesafe.pro, a free anonymous System One-compatible endpoint that forwards to TypeSafe on the operator's key.

    Serviceproduction
  • Unix filters (grev, tagv, rank, sortv, cutv, uniqv and more) that select, label and rank lines by meaning with Jev.

    CLIbetaGo
  • Portable Agent Skill teaching coding agents to use Jev as a bounded judgment layer, plus an optional MCP bridge.

    MCP serverbetaTypeScript
  • Independent agent skill that makes a coding agent design, build, test and tune a bounded Jev decision layer with a contract first.

    Agent skillbeta