What is dominicrico's jev-router?

A Claude Code plugin that automatically picks which Claude model and how much reasoning effort each prompt, step and subagent should use, so trivial work does not run at Opus prices, per the repository's own framing.

Which models does it route between?

Haiku, Sonnet and Opus per the description and badges; the badge line also lists fable. The README image says tasks are routed across those tiers with different effort levels.

The repository is named jev-router and carries a typesafe topic, and its 253 ms per pick figure is consistent with a decision-model call, but the digest excerpt truncates before any call details, so the exact integration is not evidenced here.

What guardrails ship with it?

The description lists an effort cap, a prompt-cache guard and a status band. The README's hero copy frames the goal as the right model and right effort everywhere, without paying Opus prices for trivial work.

What do the benchmarks claim?

The hero image claims minus 97% cost on trivial tasks and minus 65% on standard ones at 253 ms per pick, and notes hard tasks cost more. These are the author's own figures; the digest shows no independent verification.

The digest excerpt truncates in the links paragraph before the install section, so install steps are not evidenced here. The repository is tagged as a Claude Code mod and marketplace listing, which suggests a plugin-style install.

Who maintains it and what is its status?

dominicrico maintains it under the MIT license; GitHub metadata shows 3 stars and one fork, created 2026-10-09 and last pushed 2026-10-16, with a website at dominicrico.github.io/jev-router.

Jev directory

CommunityJev Integrations & Adapters

jev-router (dominicrico)

A Claude Code plugin that picks Haiku, Sonnet or Opus and a reasoning effort for every prompt, step and subagent, with an effort cap, a prompt-cache guard and a status band; the repo is named jev-router and tagged typesafe.

See highlights Open github.com

Stars: 3
Language: TypeScript
License: MIT
Reading time: 2 min
Published by: Community
Author: dominicrico
Last verified: Oct 2026

dominicrico/jev-router

User repository on GitHub

View on GitHub

Claude Code plugin: automatic Claude model routing. Picks Haiku, Sonnet or Opus and the reasoning effort per prompt, step and subagent; effort cap, prompt-cache guard, status band, benchmarks.

Forks: 1
Last pushed: Updated Oct 2026

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

This is one of several Jev listings for the same project. The canonical entry is Jev Router — gargpratyush.

Tagscommunity github typescript claude-code routing

Works forJev for Routing

Used inJev for Software & DevTools

Languages: TypeScript
Capabilities: Open sourceRequires API key

Highlights

Claude Code plugin that routes each prompt, step and subagent to Haiku, Sonnet or Opus with a per-task reasoning effort.
Includes an effort cap, a prompt-cache guard and a status band, with benchmarks per the repository description.
Its hero image claims minus 97% cost on trivial tasks and minus 65% on standard ones, at 253 ms per pick.
The jev-router name and the typesafe topic signal the Jev connection, but the digest excerpt shows no call details.
MIT-licensed TypeScript repo; created 2026-10-09, last pushed 2026-10-16, one fork, website at dominicrico.github.io/jev-router.

Watch out

README and GitHub metadata reviewed 2026-10-17; the excerpt truncates in the links paragraph, so install steps, configuration and the Jev integration behind the 253 ms figure are not evidenced. Cost and latency figures are the author's own claims.

Questions about this entry

Short answers drawn from the linked project or page.

Reactions & coverage

Posts, threads, and videos about this entry from around the web.

Facebook: Intake Router keeps diagnosis out of the answer space at the front desk.

FacebookClinicnotes

X: OpenRouter Decision Model Rankings

X: herdr + jev, an agent router

X: A group chat bot with decision room routing

From the Jev guides

Original write-ups on what this entry is for.

Decision gates in agent frameworks: what each integration ships

Oct 16, 2026

How Pydantic AI, Mastra, TanStack AI, DSPy, Spring AI, and LiteLLM expose Jev: where the typed decision sits in the loop, what each integration returns, and which big frameworks are only adjacent.

Jev Integrations & Adapters#agents#frameworks#integrations

Read article

From the Jev AI model community

Posts from builders shipping with Jev right now.

Follow @typesafeai