Agent tooling

jev-sift

@kbhuw46JavaScriptupdated 2026-09-18runnable

Classify first. Read selectively. A portable agent plugin and MCP tool for batch text classification.

kbhuw/jev-sift

Where it calls Jev

export const JEV_ENDPOINT = 'https://api.typesafe.ai/v1/systemone';

src/config.js:6

The link points at the commit we read, so the line number still holds.

What it asks Jev

These question sets are lifted from this project's source exactly as written.

from test/classify.test.js:10

  1. relevanttrue/false

    Relevant?

Run it on your own text

The code
import { experimental_evaluate as evaluate } from 'ai';

const { answers } = await evaluate({
  model: 'typesafe-ai/jev',
  state,
  questions: {
    relevant: { type: 'boolean', instructions: 'Relevant?' },
  },
});

from test/mcp.test.js:29

  1. healthcaretrue/false

    Serves healthcare?

  2. sectorchoice

    Pick industry

    • healthHealthcare
    • otherOther

Run it on your own text

The code
import { experimental_evaluate as evaluate } from 'ai';

const { answers } = await evaluate({
  model: 'typesafe-ai/jev',
  state,
  questions: {
    healthcare: { type: 'boolean', instructions: 'Serves healthcare?' },
    sector: { type: 'choice', instructions: 'Pick industry',
      criteria: { health: 'Healthcare', other: 'Other' } },
  },
});

Other projects in this category