A local workbench for testing your Jev questions against examples with known answers.
@DayhaysoosWrite a Jev question, add example inputs with the answers you expect, and run them. jevals saves every run so you can compare results and see whether a change made things better or worse.
Supports yes-or-no, pick-one and scored questions, and combinations of them. Starts with a single npx command and needs Node.js 22.13 or newer.
Stops a coding agent from finishing on a claim: it hooks into Claude Code and Codex, keeps a ledger of what the agent really did, and asks Jev only for the judgment calls.
Jev
TypeScript
Node.js
Claude Code
CodexA demo gate for an agent that is about to act. With no Jev key, it uses a stand-in judge and still runs.
JevTypeScriptShows a captured page crash and asks Jev who owns it, whether a contract broke, and how far it spreads.
JevTypeScript
ReactShows each visitor a fitting version of a site, and Jev updates that read as they click.
JevTypeScriptReact
VercelFlags the exact words that break a content rule, judged by Jev, at four strictness levels that do not need a new call.
Jev
JavaScriptNode.jsPuts the same code comments in front of Jev and Codex and shows what each would keep, cut, or flag.
JevTypeScript
Next.jsCodex
CodeRabbit participant
