μ · Only what's needed.
spec.jev is a coding agent. A small, fast judge called Jev makes dozens of small decisions for the main model in every turn: what stays in the context, whether a command is safe, whether the work is really done. The main model keeps its attention on the work.
$ npm i -g spec.jev
Needs Node.js 22.19 or newer. Then run spec.jev in a project folder; the first time, spec.jev setup connects a model.
v0.1.10 · The macOS app is signed and notarized by Apple; the Windows app isn't signed yet, so SmartScreen shows a warning. The app carries spec.jev inside it: no Node.js to install. Windows on Arm, Linux on Arm and every other download are under Releases.
MU · A REAL TURN
Ask for a plan, and most agents send every tool and skill description along with it. Models write better code every month, and their reports get harder to read. spec.jev splits the work: Jev judges, the main model works and a model picked for plain speaking tells you where things stand.
4 .toLowerCase() 5 .replace(/[^a-z0-9]+/g, "-") -6 .replace(/^-/, ""); +6 .replace(/^-+|-+$/g, ""); 7 } $ cd /Users/Shared/mu/slugify && npm test 2>&1 ... (10 earlier lines, ctrl+o to expand) ℹ fail 0 ℹ cancelled 0 ℹ skipped 0 ℹ todo 0 ℹ duration_ms 84.5005 Took 0.3s All 4 tests pass now. The bug was in src/slugify.js — the final .replace() only stripped a leading dash (/^-/) but not a trailing one. Changed it to /^-+|-+$/g to strip dashes from both ends. ──────────────────────────────────────────────────────────────────────────────────── ──────────────────────────────────────────────────────────────────────────────────── ~/slugify (master) ↑6.2k ↓643 R19k CH94.0% $0.034 (sub) 2.5%/200k (auto) claude-sonnet-4-6 • medium jev-1.13-free active: multi_step_task -> heavy 752ms Permissions: Full access
When you send a message
Is this a new task, a follow-up or a correction? Should a message that arrives mid-run interrupt now? Does it need skills and tools, or will a one-line answer do?
This is a real turn. "Run the tests and fix the failing ones" comes in, and in under a second Jev answers: a multi-step task, full effort. Skills and MCP servers the task doesn't need stay out of the context until it does.
When a tool answers
A test run or a search easily returns hundreds of lines. Jev judges chunk by chunk what matters now; only that enters the context, and the rest is archived behind a pointer, ready when it's needed.
In web pages and MCP results, instructions aimed at the model are withheld before it reads them.
Before acting
Before a command runs, rules flag anything risky, and Jev checks that you asked for it and that it doesn't cross a constraint you stated. There are three permission modes, Full access, Jev approves and Ask first, and you can switch at any time. In Jev approves mode, Jev lets through what the task clearly needs and asks you about the rest.
When the turn ends
When the model says it's done, spec.jev checks whether anything verified that, whether the turn drifted and whether to go back to a checkpoint.
/goal <condition> keeps the agent working until the condition holds. The main model checks whether it does, with Jev as the fallback; when the same error comes back again and again, Jev judges whether the approach is a dead end.
The hive
When several agents work on one hard problem, splitting up the work is easy; communication is hard. Say nothing, and each walks into the same dead end. Say everything, and every context fills with the others' chatter.
In spec.jev's hive, Jev judges every finding: is it worth sharing, and does it matter to this bee? Only then does it reach another bee as context. When a later finding overturns an earlier one, the correction goes to every bee that heard it.
The plain-language board
Models get stronger, and their reports read more and more like notes for a machine. In spec.jev the working model keeps working its own way. Every few tool calls Jev checks what happened, picks what's really news and hands it to a model chosen for speaking plainly, which tells you where things stand.
When a run ends, you see not just "done" but what was actually done.
WHAT WE LEFT OUT
A model's attention is limited, and so is its context. These are the things spec.jev chose not to do, or chose to do differently.
In Jev approves mode, Jev lets through what the task clearly needs and asks you about the rest.
Each message is judged: small talk, a follow-up or work to do. Skills and MCP servers that aren't needed stay out of the context.
When the context grows, stale tool results are dropped instead of summarized, and summary-free compaction is an option. Before the cache expires, spec.jev judges whether a refresh is worth it.
40 decision points, each with its own switch and its own judge: the free Jev, the local Laya or any LLM.
Without a judge key, the free Jev on OpenCode Zen answers, and spec.jev says so once a day.
The command line is built on pi, with its models, extensions, skills and session tree; the desktop app is built on AionUi.
ALSO
/status, spec.jev ledger and the app's Judge tab.GET INVOLVED
Report a bug or ask for a feature; the forms ask for what is needed.
Questions, ideas and what you built with spec.jev.
Fix a bug, propose a decision point, measure a judge or translate; good first issues are labeled.
spec.jev works without a key. To use your own TypeSafe key or another judge, see Judges.
A star helps more people find it. The code, the discussions and every release live in the repository.