Organization repository on GitHub
OpenJev: An Opencode plugin that replaces text-generation decisions with Jev (TypeSafe System One).
- Forks
- 0
- Last pushed
- Updated Sep 2026
Repo stats from the GitHub API, cached Oct 2026.
Jev directory
CommunityJev Integrations & Adapters
OpenJev for opencode
An opencode plugin that turns an agent's small decisions — is this command safe, which tool fits, is this PR ready — into typed Jev questions with calibrated confidences, labelling each answer auto or escalate so the agent knows when to ask for help.
- Stars
- 11
- Language
- JavaScript
- License
- MIT
- Reading time
- 2 min
- Published by
- Community
- Author
- darwintechlab
- Last verified
- Oct 2026
- Languages
- JavaScript
- Platforms
- CLI
- Capabilities
- Open sourceRequires API key
Highlights
- Adds five tools: jev_choice, jev_noul, jev_score, jev_ask and jev_doctor, plus an optional skill.
- The repo measured p50 267 ms per decision and about $0.017 per 1,000 decisions, with no output-token cost.
- In its tests, pick-one answers scored 0.75 or higher were right 96% of the time; answers are labelled auto or escalate.
- Up to 32 questions per call; 27 questions took about 5 ms longer than 1 in the reported tests.
- MIT, Node >= 20, opencode >= 1.18; it never writes code and needs a fixed set of answers.
Watch out
Grounded on a truncated README and GitHub metadata captured 2026-10-15; the digest cuts off in the 'How it works in opencode' section, so setup specifics beyond the 5-minute SETUP.md pointer are not covered here. The speed, cost and accuracy numbers are the project's own measurements, repeated as published.
Questions about this entry
Short answers drawn from the linked project or page.
A plugin that gives the opencode agent a decision model for the small questions it faces constantly — is this command safe, which tool fits, is this PR ready, how serious is this bug — answered as typed choices with calibrated confidence instead of generated text.
Installing the plugin adds five tools: jev_choice, jev_noul, jev_score, jev_ask and jev_doctor. If you also turn on the bundled skill, the agent uses the tools by default without being asked.
Every answer is labelled auto (confident enough to act on) or escalate (should go to the LLM or a person). The project's tests found pick-one answers it scored 0.75 or higher were right 96% of the time.
The repo reports p50 267 ms per decision and about $0.017 per 1,000 decisions, with no charge for output tokens. Up to 32 questions can ride in one call; 27 questions took about 5 ms longer than 1 in its tests.
It does not write code, summaries or explanations, and it cannot answer questions that lack a fixed set of options. The README positions it alongside the LLM, handling small decisions so the LLM does not have to.
MIT-licensed, Node 20 or newer and opencode 1.18 or newer, plus a key for the hosted Jev decision model. The repo points to a five-minute setup guide for installation.
darwintechlab, with the README crediting darwintechlab.com, darwinevo.com and the community. The digest's README is truncated, so release history and support commitments cannot be checked from it.
More Jev resources like this
A Claude Code plugin that adds five Jev tools — jev_choice, jev_noul, jev_score, jev_ask and jev_doctor — so small decisions like safety checks, tool choices and review verdicts come back as typed answers with calibrated confidence, labelled auto or escalate.
Jev Integrations & Adapters#community#catalog-import#logicrw
Communitydarwintechlab/claude-jevJavaScript
darwintechlab/claude-jevJavaScriptAn OpenCode plugin that asks Jev how much reasoning each step needs — low for a rename, high for a failing test — and adds the picked effort to the request, falling back to high when Jev is slow or unavailable. Archived and superseded by reasoning-router.
Jev Integrations & Adapters#community#catalog-import#logicrw
An OpenCode plugin that keeps the session on a cheap sticky parent model so the cache stays warm, and asks Jev whether a turn is hard before escalating through jev_escalate to temporary stronger subagents.
Jev Integrations & Adapters#community#catalog-import#cobanov
A local Codex plugin for codebase search: it runs ripgrep, asks Jev a binary Yes/No relevance question for each candidate passage, and returns those above a probability threshold (default 0.50). Codex can expand results or surrounding context without extra Jev calls.
Jev Integrations & Adapters#community#catalog-import#logicrw
Community
zbush/jev-contextJavaScript
zbush/jev-contextJavaScriptRoute OpenAI Codex CLI turns through Jev: a local loopback Responses API proxy asks Jev to pick a Codex model and reasoning effort for each fresh turn, then forwards the request. If Jev is unavailable, Codex continues unchanged.
Jev Integrations & Adapters#community#catalog-import#madewithjev
Community
tiandee/codex-jev-routerJavaScript
tiandee/codex-jev-routerJavaScriptLocal tool-routing classifier for coding agents: a gateway and MCP tool that ask Jev which tool to call next, keep your agent's own model and login, and log every decision so you can start by observing.
Jev Integrations & Adapters#community#catalog-import#logicrw
Community
felpsdev/jev-classifierTypeScript
felpsdev/jev-classifierTypeScriptFrom the Jev guides
Original write-ups on what this entry is for.
From the Jev AI model community
Posts from builders shipping with Jev right now.