What is jev-pilot and what does it do?

A Claude Code plugin that asks Jev a few typed questions about your prompt before every turn and sets the turn up from the answers, so easy work runs at low effort on cheaper subagents and hard work gets the thinking it needs.

What does Jev decide on each turn?

Reasoning effort from low to xhigh at the start of each turn, subagent model and effort when a subagent starts, strategy advice, and the one skill the prompt needs. It can also raise effort to max once mid-turn when tool calls keep failing.

How do I install jev-pilot?

Run the install script with curl, which asks for your OpenRouter key, installs jev-pilot as a regular Claude Code plugin, and adds the claude-jev command. Then start Claude Code with claude-jev.

What does running jev-pilot cost?

Jev on OpenRouter costs about $0.04 per million input tokens with free output. You need your own OpenRouter key, and provider billing applies.

What are the requirements?

Claude Code 2.1.278 or newer with function hooks, which are early access and need CLAUDE_CODE_ENABLE_FUNCTION_HOOKS=1; the claude-jev launcher sets that for you.

Does Jev write in my conversation?

No. Jev talks through Claude the pilot, a small animated pet above the prompt, while you keep Opus for the conversation. The /jev-pilot:report command records every decision with tuning suggestions.

What is the license and how mature is it?

The README badge says MIT while GitHub metadata reports NOASSERTION. Function hooks are early access, and the README was truncated after installation in the digest, so report and configuration details could not be reviewed.

Jev directory

CommunityJev Agents & MCP

Jev Pilot — akramovic1

Claude Code plugin that asks Jev a few typed questions before every turn and sets reasoning effort, subagent model and effort, strategy, and the skill a prompt needs; you keep Opus for the conversation.

See highlights Open github.com

Stars: 12
Language: TypeScript
Reading time: 2 min
Published by: Community
Author: Akramovic1
Last verified: Oct 2026

akramovic1/jev-pilot

User repository on GitHub

View on GitHub

Let Jev steer Claude Code: the right reasoning effort, subagent model and skill for every prompt. A Claude Code plugin powered by TypeSafe's Jev (OpenRouter / TypeSafe).

Forks: 2
Last pushed: Updated Sep 2026

Repo stats from the GitHub API, cached Oct 2026.

Tagscommunity catalog-import logicrw claude-code routing openrouter

Works forJev for Routing Jev for Decision gating

Used inJev for Software & DevTools

Languages: TypeScript
Platforms: CLI OpenRouter
Capabilities: Open sourceRequires API key

Highlights

Before each turn Jev sets reasoning effort from low to xhigh, and mid-turn it can raise effort to max once when tool calls keep failing.
It also decides subagent model and effort (Haiku, Sonnet, or Opus), strategy advice, and the one skill a prompt needs, if any.
Jev never writes in your conversation: a small animated pet above the prompt shows what Claude is doing and what Jev decided.
The installer asks for your OpenRouter key and adds a claude-jev launcher; Jev costs about $0.04 per million input tokens with free output.
Needs Claude Code 2.1.278+ with early-access function hooks and CLAUDE_CODE_ENABLE_FUNCTION_HOOKS=1, which claude-jev sets for you.

Quickstart

bash

curl -fsSL https://raw.githubusercontent.com/Akramovic1/jev-pilot/main/install.sh | bash
claude-jev

Watch out

Copy reviewed against the repository README on 2026-10-15; it was truncated after the install section, so the report, configuration, and acknowledgements sections were not inspected. GitHub metadata reports license NOASSERTION while the README badge says MIT, and the illustrated session's numbers are examples rather than measurements.

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

Other listings of this project

Forks and copies of this project in the directory. This entry is the canonical one.

jev-pilot

Python helper that wraps TypeSafe Jev for agent arbitration, pre-execution guard, stuck-loop checks, fact verification and intent routing with probabilities and confidence.

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

Community

h0j5bz0adh0-stack/jev-pilotPython

From the Jev guides

Original write-ups on what this entry is for.

MCP tool routing: Jev in front of the tool call

Oct 16, 2026

Tool menus can outgrow the context that reads them. The catalog's Jev-based routers and gates narrow the list: benchmarked tool search, per-tool scoring, fail-closed approvals, and WebMCP in the browser.

Jev Agents & MCP#mcp#tool-routing#agents

Read article

Approval policies for coding agents: hooks, guards, and gateways