How does Jev decide whether an automation should fire?

Each tick runs an observe step, then the jev_when tool sends a single Boolean Noul decision with your when condition, not-when condition, evidence and optionally the previous observation through OpenClaw's decisionModel role. Below the 0.7 threshold it returns fire:false and no agent turn runs.

Can it run without the hosted Jev API?

Yes. The decisionModel can point at a local Kev server or an OpenClaw ONNX CPU classifier without changing the plugin or generated scripts. The README reports those offline options scored lower (83% and 54%) on its synthetic tick set than hosted Jev's 99%.

What does a quiet tick actually cost?

Nothing but the judge call. When the answer is no the automation returns fire:false and no conversation model runs. The README puts hosted Jev at about $0.000016 per check, so a 5-minute watcher costs roughly $0.14 a month.

What is the difference between edge and level firing?

By default the trigger fires only when the condition becomes true, once, rather than every tick while it stays true. --level switches to firing on every matching tick, which suits a price-drop watcher.

How does it treat a judge that is simply unreachable?

Unavailable is not treated as no, so a broken watcher cannot look healthy: the script stays quiet but fires one alert after three consecutive misses by default. --alert-after n changes that count, and 0 disables the alert.

What does --compare add to a trigger script?

It passes the last run's observation to jev_when, so conditions like dropped, changed or went up can be judged against the previous state. --keep head controls which end of long evidence is cut.

What do I need installed to use this plugin?

OpenClaw 2026.9.6 or later, the release that shipped the decisionModel role, and Node 24.16 or newer. The judge is whatever the decisionModel points at, hosted Jev by default, or a local Kev server or ONNX classifier.

CommunityJev Tools & CLIs

OpenClaw Jev Trigger

OpenClaw plugin and CLI that turn a plain-language --when/--not-when condition into a scheduled trigger script. Each tick sends one Noul through OpenClaw's decisionModel and only wakes the conversation model when the condition reaches 0.7 or above; a 76-tick benchmark reports 75 correct judgments and 0 false wake-ups.

See highlights

Open github.com

Stars: 1
Language: TypeScript
License: MIT
Published by: Community
Last verified: Oct 2026

yousan/openclaw-jev-trigger

User repository on GitHub

View on GitHub

Jev × OpenClaw: write automation triggers in plain language. Jev (TypeSafe) judges every tick through OpenClaw's decisionModel role; the model only wakes when it matters.

Forks: 1
Last pushed: Updated Sep 2026

Repo stats from the GitHub API, cached Oct 2026.

About this project

Write your automation's trigger in plain language. Jev decides, every tick, whether it's time to wake the model. openclaw-jev-trigger is an OpenClaw plugin. It adds one tool, jevwhen, that asks the agent's decision model whether a condition written in ordinary words is true for what a watcher just observed. It also ships a small CLI that writes the trigger script for you.

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

Tagscommunity automation openclaw triggers

Works forJev for Decision gating

Used inJev for Software & DevTools

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

Highlights

Generates trigger scripts from flags like --when, --not-when, --exec, --fetch and --compare.
76 synthetic ticks: hosted Jev scored 75 correct with 0 false wake-ups at 231ms p50, vs 87% for first-try JavaScript rules.
Edge-triggered by default so conditions fire once, not every tick; --level switches to repeated firing.
Judge failure never counts as no: the script stays quiet and alerts after three consecutive misses.

Watch out

MIT licensed. Benchmark data is synthetic and written by the same AI agent that built the plugin, so the README calls it a smoke test rather than a leaderboard; hosted Jev sends observations to TypeSafe, so private data should use Kev or ONNX judges instead.

Questions about this entry

Short answers drawn from the linked project or page.

More Jev resources like this

Oh My ClaudeCode

39.8k

A teams-first multi-agent orchestration plugin for Claude Code with opt-in Jev hooks for sub-millisecond judgment points, decision caching and per-point egress controls. TypeScript, MIT, with a staged team pipeline and tmux CLI workers.

Jev Tools & CLIs#community#claude-code#orchestration

Community

Yeachan-Heo/oh-my-claudecodeTypeScript

GenOffice

9.2k

An open-source AI office suite whose local file search reranks the top 20 FTS hits with TypeSafe Jev relevance scores before showing results.

Jev Tools & CLIs#community#typescript#desktop

Community

genspark-ai/genofficeTypeScript

3.4k

Vercel Labs' Unix-like coding agent CLI written in Zig, shipping a typesafe_permission_reviewer builtin that runs the agent's permission decisions through Jev rather than an LLM call. Model-agnostic and embeddable via libfx and ACP.

Jev Tools & CLIs#community#coding-agents#cli

Community

vercel-labs/fxZig

jevgrep (code search)

2.6k

A CLI that asks Jev which files and declarations answer a repository question, then prints verbatim excerpts for a coding agent to use.

Jev Tools & CLIs#community#typescript#cli

Community

dzhng/jevgrepTypeScript

surf

634

A CLI browser automation tool with optional Jev semantic mode: Surf observes, Jev selects the next action, Surf validates and acts, Jev checks the goal.

Jev Tools & CLIs#community#browser#cli

Community

nicobailon/surf-cliTypeScript

Jev-Code-Reviewer

140

A local Chrome extension plus CLI that classifies each agent PR hunk as P0, P1, or P2 with Jev, showing P0 by default and human-readable diff summaries.

Jev Tools & CLIs#community#code-review#typescript

Community

egma-ai/jev-code-reviewerJavaScript

From the Jev guides

Original write-ups on what this entry is for.

Jev on the Gateways: Vercel, Workers AI, and OpenRouter

Oct 12, 2026

The request shape barely changes across surfaces; the surrounding contract does. When to call the API directly and when a gateway, a Worker binding, or a router entry earns its extra hop.

Jev Tools & CLIs#gateway#vercel#cloudflare

Read article

Exporting Decision Models: GGUF, ONNX, and Device Runtimes

Oct 10, 2026

Taking a decision checkpoint out of Python: what the GGUF and browser WebGPU exports preserve, what they change, and the parity evidence the model cards publish.

Jev Tools & CLIs#open-models#local#onnx

Read article

Records at Rest: Documents, Inboxes, and Risk Queues

Sep 28, 2026

Jev labels a record that already exists. Code moves, flags, or holds it. How the document, mail, and transfer tools keep the action out of the model.

Jev Tools & CLIs#classification#email#security

Read article

Back to all resources

From the Jev community

Posts from builders shipping with Jev right now.

Follow @typesafeai