A local gateway that lets Jev pick which tool your coding agent calls next, and leaves everything else alone.
@oviniciuslanaCoding agents spend a lot of expensive reasoning just deciding which tool to use next. This gateway sits between your agent and its model and, at exactly that moment, asks Jev instead.
Every other request passes through to your usual model untouched. It works with Codex, Claude Code and OpenCode out of the box, including on ChatGPT and claude.ai subscriptions.
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 code checker that asks Jev the questions a parser cannot, like whether a comment, a name or a test actually matches the code it sits on.
JevTypeScriptFind code by asking what it does: a CLI that uses Jev to judge relevance across a codebase instead of matching patterns.
JevTypeScriptNode.jsA pre-execution guard that sits between a terminal coding agent and your shell, asking Jev how likely a command is to destroy something before it runs.
JevTypeScriptTurn your Jev usage into small, very fast classifiers that run locally in Node.js or the browser.
JevTypeScriptNode.jsA second opinion for coding agents in half a second: a Jev command-line tool that can check every command first.
JevTypeScript
