What does calling a recipe actually return to my code?

Each recipe takes your data, calls Jev through TypeSafe or an injected compatible client, and returns a structured decision while your app decides what happens next. The route example returns a team name on ready and a review path when uncertain.

How should I evaluate recipes before wiring a live key?

Start with route, rerank or tool-call-gate guides, try npx demo variants on saved responses, then use the getting-started workflows and support-routing starter for clarify-and-resume flows. Live calls need a TypeSafe key via direct or Gateway integration.

Will it fit my existing Vercel or LangChain stack?

It plugs into Vercel AI SDK or LangChain or runs standalone, with the same recipe interface for direct and Gateway paths. Recipes run in Node backends, scripts or research evals and need Node 22.9 or newer with ES modules.

How do I know which evidence behind a recipe to trust?

The catalog site, fixture exploration and retained evaluation evidence label measurement origin explicitly, and the installed evaluator lets you replay confidence policies offline. Check version tags and release notes since hosted docs can include unreleased changes.

What runtime and license does it need?

Node.js 22.9 or newer, ES modules, and a TypeSafe API key for live calls. It is MIT licensed and runs in a Node backend, a script, or a research evaluation.

Can I try a recipe without a key?

Yes, some demos use saved responses: npx jev-recipes demo route, demo rerank, and demo tool-call-gate run without a TypeSafe key, and the workflow examples show ready decisions, uncertainty, and provider failures.

CommunityJev Agents & MCP

Jev Recipes

Q: How many recipes are there and what do they cover?

248 focused JavaScript and TypeScript recipes that route messages, check evidence, and label model responses with a function call. Each returns a structured decision, and your application decides what happens next.

TypeScript recipe library with 248 focused functions that call Jev for routing, reranking, gating and labelling, returning structured ready-or-review decisions for Node backends.

See highlights

Open github.com

Stars: 20
Language: TypeScript
License: MIT
Published by: Community
Author: agencyenterprise
Last verified: Oct 2026

agencyenterprise/jev-recipes

Organization repository on GitHub

View on GitHub

248 small Jev decisions for JavaScript and TypeScript: route, rerank, gate, grade, compare, and label text for agents, RAG, support, code review, and music. Each returns a typed result with explicit uncertainty. Plug into Vercel AI SDK or LangChain agents, import one recipe, or run the CLI with JSON from any language.

Forks: 1
Last pushed: Updated Oct 2026

Repo stats from the GitHub API, cached Oct 2026.Homepage

About this project

Typed AI decisions for agents, RAG, and support flows: route, rerank, gate, grade, compare, and label text with one function call. Plug into Vercel AI SDK or LangChain, or call any recipe directly. Recipe guides at jev-recipes.com | Quickstart | Live music app | Coding assistants | Recipe catalog | API vs. SDK vs. recipes | Agent frameworks | Contributing. 248 focused recipes for JavaScript and TypeScript. Route messages, check evidence, and label model responses with a function call.

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

Tagscommunity awesomejev catalog-import

Works forJev for Classification Jev for Evaluation Jev for Tool calling

Used inJev for Software & DevTools

Languages: TypeScript

Highlights

Route, rerank and tool-call-gate starters with saved-response demos and workflow examples.
Uncertainty-aware results that propose a queue or request human review when unresolved.
Framework adapters plus direct and Gateway integrations under one recipe interface.
Support-routing starter and static catalog with evaluation evidence and audit command.

Watch out

Demos use saved responses and the music app is a maintainer-built example, not an independent adoption claim or accuracy benchmark, so use live calls and retained evals for decisions. 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

construct-auto-classifier

Shell gate for coding assistants that judges commands by effect and blast radius, asking Jev twelve questions or a chat model, with fast rules first.

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

Community

godspede/construct-auto-classifierTypeScript

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.