Does it need training data or labels?

No. You type a statement such as "The customer is angry" and Jev scores each row zero-shot; the filter keeps rows at or above the threshold, 0.8 by default.

What happens on large tables?

When more rows than maxUnscoredRows (1,000 by default) lack cached scores, the filter pauses and shows a Run anyway button, and you can narrow the table with other filters or a search first.

What versions of Laravel and Filament does Vibefilter require?

PHP 8.2+, Laravel 12.36+ or 13, and Filament 5.9+. Earlier Filament 5 releases will probably work but are not supported because the test suite cannot run against them; composer update filament/filament gets an older 5.x there.

How do I install Vibefilter, and where does the API key go?

Run composer require vibefilter/filament, publish and run the migration with php artisan vendor:publish --tag="vibefilter-migrations" then php artisan migrate, and set TYPESAFE_API_KEY in .env. The migration creates two tables for the score cache.

Can I use Jev through OpenRouter instead of TypeSafe?

Yes. Set VIBEFILTER_DRIVER=openrouter and OPENROUTER_API_KEY; the model becomes typesafe/jev-1.13, and OpenRouter forwards to TypeSafe. Both drivers are pinned to a fixed model version rather than an alias like jev-latest, because a silent update would mix scores from two versions in the cache.

How does Vibefilter behave when the API errors or rate limits?

Requests that hit a rate limit, a server error or a dropped connection are retried after 0.5, 2 and 5 seconds, honouring a Retry-After header up to 30 seconds; a batch that still fails is split in half and tried once more. Every returned batch is saved immediately, so the table filters on scored rows and a warning offers to try the rest.

Can I show users how sure the model is about each row?

Add the optional VibeScoreColumn; it shows each row's probability from the run the filter already did, asks the model nothing and needs no column in your database. It is not sortable, and rows waiting for Run anyway show a dash. Most useful while choosing a threshold.

CommunityJev Integrations & Adapters

Vibefilter

Filament plugin that filters admin table rows by a plain-English statement: each row gets a Jev Noul probability that the statement is true, and rows at or above the threshold stay. Scores are cached by content, so reruns of the same statement are instant and free.

See highlights

Open github.com

Stars: 4
Language: PHP
License: MIT
Published by: Community
Last verified: Oct 2026

vibefilter/filament

Organization repository on GitHub

View on GitHub

Filter Filament tables with a plain-English statement. Each row is scored by TypeSafe Jev.

Forks: 1
Last pushed: Updated Oct 2026

Repo stats from the GitHub API, cached Oct 2026.Homepage

About this project

Filtering beyond SQL. Filter your Filament tables by vibe. Try the live demo → · vibefilter.dev. Vibefilter turns any sentence into a zero-shot classifier for your Filament table: each row gets a calibrated probability that the statement is true, and the table keeps the rows above your threshold.

Quoted from the project's README. Read the full README.

Tagscommunity laravel filament php

Works forJev for Classification Jev for Scoring Jev for Decision gating

Used inJev for Customer Support Jev for E-commerce & Retail

Languages: PHP
Platforms: OpenRouter Browser
Capabilities: Open sourceRequires API key

Highlights

On the 1,000-review demo, a 0.8 threshold matched the reviews' own mood labels on 861 rows.
Reads rows in batches of 100 across 10 parallel requests; a run over 1,000 short reviews costs about $0.003 and a second.
Scores cache against a hash of row text, statement and driver, keyed by content rather than row ID.
Requires PHP 8.2+, Laravel 12.36+ and Filament 5.9+; ships a FakeDriver so tests never call the API.

Watch out

MIT licensed. The text columns you pass are sent to TypeSafe or OpenRouter, and the demo results measure agreement with model-generated labels, not ground truth.

Questions about this entry

Short answers drawn from the linked project or page.

More Jev resources like this

AutoGPT

187.5k

Autonomous agent platform with 187,504 stars for building and running agents from plain English or a visual builder. Its platform features first-class TypeSafe Jev decision blocks for typed routing, filtering, scoring and confidence-gated next-action dispatching.

Jev Integrations & Adapters#community#autonomous-agents#platform

Community

Significant-Gravitas/AutoGPTPython

Inbox Zero

12.4k

Open-source AI email assistant (Next.js, Postgres, 12k+ stars) with a decision-model layer that can run on openrouter:typesafe/jev-1.13: Jev-backed questions cover rule selection, cold-email detection, thread status, recurring patterns, unsubscribe pages, and sender categorization, each with evals.

Jev Integrations & Adapters#community#email#gmail

Community

elie222/inbox-zeroTypeScript

IntentSQL

Alpha playground that answers plain-English questions over SQLite by composing small Jev decisions -- table, column, comparison and operand selection -- into inspectable typed plans, deterministic SQL and previewed guarded writes. Bundles Cyberchase, DESE and Moneyball databases.

Jev Integrations & Adapters#community#sqlite#text-to-sql

Community

Amine-LG/IntentSQLPython

Rot Guard

MIT-licensed Chrome extension that asks Jev whether the page, video or post you are on serves your written goals -- a category Choice, a goals Noul and a 0-4 waste-of-time Score -- then blocks or blurs in code, tightening thresholds at 2, 10 and 20 minutes of engaged time.

Jev Integrations & Adapters#community#chrome-extension#digital-wellbeing

Community

plusminushalf/rot-guardJavaScript

Pick Verify

A warehouse pick-verification step that compares a scanned item, the bin's expected contents, and the order line to decide between accept and re-scan, catching confusable picks a barcode alone would pass.

Jev Integrations & Adapters#community#python#robotics

Community

SAP S/4HANA Procurement Fraud

Proof of concept that flags suspicious purchase orders in SAP S/4HANA: ABAP computes risk indicators, a Fiori Elements app sends seven typed questions to Jev via OpenRouter, and ABAP applies thresholds to rate each order. Runs in shadow mode only.

Jev Integrations & Adapters#community#sap#fraud-detection

Community

seccoit/sap-s4-antifraude-jevJavaScript

Back to all resources

From the Jev community

Posts from builders shipping with Jev right now.

Follow @typesafeai