Organization repository on GitHub
grep that reads meaning. Make your agent's monitor smart: Jev judges each line, and only what matters wakes the agent.
- Forks
- 2
- Last pushed
- Updated Oct 2026
Repo stats from the GitHub API, cached Oct 2026.Homepage
Jev directory
CommunityJev Tools & CLIs
Jevable
A bash-native CLI from Parall that gives an agent Jev as an atomic tool: pipe any event stream through a plain-English question and it prints only the lines that pass, composing with the tools your agent already pipes.
- Stars
- 2
- Language
- TypeScript
- License
- MIT
- Reading time
- 3 min
- Published by
- Community
- Author
- parall-hq
- Last verified
- Oct 2026
Works forJev for Classification
Used inJev for Software & DevTools
- Languages
- TypeScript
- Platforms
- CLIOpenRouterVercel AI Gateway
- Capabilities
- Open sourceRequires API key
Highlights
- Answers follow grep exit codes: 0 for a match, 1 for none, 2 for an error; --json adds scores, -t 0.8 tightens the threshold and -v inverts.
- A 334-event benchmark in the README reports 86% lower always-on cost than a 30-minute heartbeat while catching 85% of what mattered.
- Rules can mix ordinary conditions with Jev judgments; setup is npx -y jevable guide, which connects a key, tries a filter and sets up a watch.
- Persistent watches use --key to identify repeated events, --state to remember them, --cooldown to limit repeats and --window to judge whole intervals.
- Requires Node.js 20.3+ and a key from TypeSafe, OpenRouter or Vercel AI Gateway; matching lines pass through unchanged.
Quickstart
bash
npx -y jevable key YOUR_API_KEY
tail -n 0 -F app.log |
npx -y jevable "Does this line report that a dependency is down?"Watch out
Copy reviewed against the truncated repo README on 2026-10-15; the excerpt cuts off inside the watching guidance. The 86% cost-cut and 85% catch figures are the project's own 334-event benchmark, and the README publishes no pricing of its own beyond the key sources.
Questions about this entry
Short answers drawn from the linked project or page.
It filters any stream by meaning: you pipe events in with a plain-English question and it prints only the lines that pass, unchanged, ready for the next command. The README calls it giving your agent Jev as an atomic tool, built for monitors and watches.
It behaves like grep: exit 0 for a match, 1 for no matches, 2 for an error, with matching lines passed through unchanged. --json includes scores, -t sets a stricter threshold and -v inverts the filter, and it composes with tools like tail, gh and jq.
In the README's 334-event benchmark, jevable cut always-on cost by 86% compared with a 30-minute heartbeat while catching 85% of what mattered. Those are the project's own numbers from its benchmark page, not an external evaluation.
Node.js 20.3+ and a Jev API key from TypeSafe, OpenRouter or Vercel AI Gateway, saved once with npx -y jevable key YOUR_API_KEY. The README suggests pasting npx -y jevable guide into your agent to have it set everything up.
Agents that run shell commands, including Claude Code, Codex, OpenClaw, Hermes, pi and dsh; the guide covers how each runtime receives matches. Anything that can pipe a stream can use it without an agent framework.
Yes: --key identifies repeated events, --state remembers them across runs, --cooldown limits repeat notifications, and --window judges a whole interval, including volume spikes or a missing heartbeat. Watch sources include logs, issues, PR comments, feeds and releases.
It judges each line with Jev rather than sending the whole stream to a generative model, so matching lines are passed through unchanged and your question decides only what passes. --on .body judges one field via a jq-style path while keeping the whole record.
More Jev resources like this
An npm CLI that scans TypeScript, JavaScript and Python codebases for LLM calls that are really decisions, converts them into Jev modules with confidence-gated fallbacks, and shadow-tests them on recorded traffic before you swap.
Jev Tools & CLIs#community#catalog-import#logicrw
Community
0xjba/jev-swapHTML
0xjba/jev-swapHTMLWrite end-to-end tests for web apps in plain English; jev-e2e turns each case into a plan, has Jev pick the page controls, and runs it with Playwright, returning PASS, FAIL or BLOCKED with HTML, JSON and masked screenshot evidence.
Jev Tools & CLIs#community#catalog-import#madewithjev
Community
perixtar/jev-e2eTypeScript
perixtar/jev-e2eTypeScriptA Chrome extension that reorders the Gmail message list on screen only - unread first, critical messages on top - using Jev to score each row's urgency and criticality and to answer one boolean question per custom label.
Jev Tools & CLIs#community#catalog-import#madewithjev
Community
iamomiid/jev-inboxTypeScript
iamomiid/jev-inboxTypeScriptModeration for communities and apps, powered by Jev: every message gets a probability for spam, scam, harassment, nsfw, doxxing and more, plus rules in plain English. Thresholds and actions are yours, and flag-only is the default.
Jev Tools & CLIs#community#catalog-import#logicrw
Community
ohernandezdev/jevmodPython
ohernandezdev/jevmodPythonA stateless Rust CLI for typed semantic judgments with the Jev model: state and questions in, calibrated answers and exit codes out, as stable JSON on stdout. Ships with an Agent Skill for scripts, CI and agent workflows.
Jev Tools & CLIs#community#catalog-import#madewithjev
Community
model-clis/jevRust
model-clis/jevRustA Pi coding-agent extension that uses Jev to classify each request and route it to one of three configured models. Supports TypeSafe, Cloudflare AI Gateway, Vercel AI Gateway and OpenRouter backends, with doctor checks, shadow mode and an off-by-default setup.
Jev Tools & CLIs#community#coding-agents#routing
From 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.