Why not just prompt a chat model for these decisions?

Prompts return text you parse with no sense of certainty. A jevel holds the typed questions, Jev answers with confidence, and the jevel maps that to act when sure, mark plus human flag when fairly sure, and fall_back to the old path when unsure.

How does this fit a coding agent workflow?

Store questions in ./jevels/ for the agent to read and edit while you review thresholds, then load once and call decide where code already branches. Generated TypeScript declarations keep each jevel's answer options typed in your editor.

What happens when TypeSafe is down or I have no key?

Every question returns fall_back with the reason in the error field, so code keeps its previous behavior instead of throwing. If you have no key yet, Try explains how to add one rather than failing obscurely.

Do I have to write every decision file from scratch?

Seventeen jevels ship for tickets, messages, issues, pull requests, logs, alerts and failing tests, each checked against the real API on its own cases. jevelry check refuses defective jevels and warns about questions Jev handles poorly.

How do I install jevelry?

It needs Node 22 or newer; npm install -g jevelry then jevelry install puts the skill into every coding agent it finds and asks for your TypeSafe key. Skip the key with Enter and add it later, or run npx jevelry install inside an agent.

What are the three decision outcomes?

act (Jev is sure, your code uses the answer), mark (Jev is fairly sure, use it and flag it for a person), and fall_back (Jev is unsure, do what you did before jevelry).

What license is it and how stable?

MIT, but it is early development: commands, the jevel format, and the answer document may change between minor versions until 1.0, so pin the version you depend on.

CommunityJev Agents & MCP

Jevelry

NPM package and Ink TUI for asking Jev small typed questions stored as jevel files, mapping confidence to act, mark, or fall_back decisions with every decision kept in a local log.

See highlights

Open github.com

Stars: 11
Language: TypeScript
License: MIT
Published by: Community
Author: backant-io
Last verified: Oct 2026

backant-io/jevelry

Organization repository on GitHub

View on GitHub

Use Jev everywhere to make & track decisions

Forks: 1
Last pushed: Updated Oct 2026

Repo stats from the GitHub API, cached Oct 2026.

About this project

Overview · Install · Usage · Commands · Keys · Jevels · Configuration · Jev docs. Use Jev everywhere to make & track decisions. jevelry lets your code and your coding agent ask Jev, the decision model from TypeSafe, a small typed question and get back a decision it can act on, and it keeps every decision so you can see how often Jev was right.

Quoted from the project's README. Read the full README.

Tagscommunity awesomejev catalog-import

Works forJev for Classification

Used inJev for Software & DevTools

Languages: TypeScript

Highlights

Seventeen ready-made jevels plus ask, run, outcome, report, check, types, and models commands.
Command-running jevels execute the chosen command, prompt first on mark, and support --dry-run safely.
State travels over stdin and a pointed-at file so hostile ticket text stays inert data.
Review flow records agree or disagree outcomes and reports agreement per jevel and question.

Watch out

Early development with commands, jevel format, and answer documents subject to change between minors until 1.0, so pin the version you depend on. Copy reviewed against the repository README on 2026-10-04, not from running it.

Questions about this entry

Short answers drawn from the linked project or page.

More Jev resources like this

jev-code

TypeScript SDK and CLI coding agent where Jev only picks options while your code runs effects, building programs as syntax trees.

Jev Agents & MCP#community#awesomejev#catalog-import

Community

rhighs/jev-codeTypeScript

JevLoop

Runnable agent harness that moves pick, score and yes-no loop choices to Jev or Laya in milliseconds while reserving the language model for writing, compiling DECISION.md into executable questions.

Jev Agents & MCP#community#jevbest#catalog-import

Community

zjunlp/JevLoopTypeScript

jev-test-filter

Diff-aware test selector that sends the git diff to Jev with one question per test, then prints the native filter flags and hands them to vitest, jest, node, bun, Playwright, cargo, or go test.

Jev Agents & MCP#community#jevbest#catalog-import

Community

mizchi/jev-test-filterTypeScript

pi-jev-router

Pi extension that classifies each task with Jev over OpenRouter and routes it to a Pareto-optimal model balancing measured quality, cost, and latency under a role policy.

Jev Agents & MCP#community#jevbest#catalog-import

Community

philippdubach/pi-jev-routerTypeScript

patdown

Node CLI that lints a file tree against fuzzy natural-language rules in one Markdown file, judging each matched file with a swappable TypeSafe-backed judge.

Jev Agents & MCP#community#jevbest#catalog-import

Community

tyler-dot-earth/patdownTypeScript

jevcore

106

Decision layer exposing Jev's typed yes-no, choice, and score answers to DeepSeek Harness and MCP hosts, offline by default with every network transmission disclosed per feature.

Jev Agents & MCP#community#jevbest#catalog-import

Community

PerryLink/jevcoreTypeScript

Back to all resources

From the Jev community

Posts from builders shipping with Jev right now.

Follow @typesafeai