GitHub radar

Let a cheap AI sort routine decisions before Claude reads anything

This free Claude Code skill puts a tiny decision model in front of Claude: it sorts emails, leads, or comments in a fraction of a second, and Claude only steps in when it's unsure.

Jev is a tiny model that answers yes/no or picks one option from a list — it never writes sentences. This skill wires it into Claude Code as a first filter before Claude.

When it helps

Useful if dozens of similar decisions flow through your agent every day — sorting emails, scoring leads, checking a draft before sending. Skip it for one-off tasks or anything that needs context and explanation, not just a label.

Pros

  • Answers in about a third of a second for a fraction of a cent
  • Reports a confidence score with every answer
  • Can't invent facts — it has no words to make them up with
  • Frees up Claude's attention for genuinely unclear cases

Cons

  • Can't write text or explain its own decisions
  • Doesn't remember context or connect separate answers
  • Needs a paid OpenRouter key, even if a cheap one

How to set it up — step by step

  1. 1Open your AI agent (Claude Code) in the folder of the project where you want to sort emails or leads.
  2. 2Send the agent the text from the block below — it will install the skill, ask for an OpenRouter key, and run a test call.
  3. 3Give the agent your OpenRouter key if it asks for one (get it at openrouter.ai, under Settings → Keys).
  4. 4Check the summary line it prints: how many items, how many seconds, and what it cost.
  5. 5If you like the result, ask the agent to turn on /jev-route so Claude only reads the unsure cases.

Text for your agent

Copy this and send it to your agent — Claude Code, Codex, any of them:

Install the Claude x Jev skill for Claude Code from https://github.com/charlesdove977/claude-x-jev - follow the README (npx claude-x-jev install --with-commands), set up an OpenRouter key (ask me for it) and run /jev-classify on a couple of my test emails. Show me the summary line: how many items, how many seconds, and what it cost.

Light: a regular laptop with Python 3.8+, no GPU needed. But the service itself is not free - you need your own OpenRouter key, though a cheap one: the author says five dollars of credit covers hundreds of thousands of decisions.

Open on GitHub