Skip to content
Metatext Decide — typed decisions for your agents, with a probability on every answerGet early access
Metatext
Sign Up

Jev Web Analyzer

Community tool that fetches a SaaS homepage as Markdown and asks Jev ten questions about first-visit clarity, with a leaderboard.

ServicebetaApache-2.0
View on GitHub ↗Live demo ↗

Overview

Jev Web Analyzer lets you paste a public SaaS homepage and see what a first-time visitor, and Jev, can tell about it. ReplyNodes fetches the page and extracts clean Markdown, and Jev, called through Vercel AI Gateway, answers ten bounded typed questions over that shared state about audience, clarity, differentiation, call to action, trust and product motion, including the first change to make. URL validation, caching, sanitisation and presentation are deterministic.

The author's stated purpose is to make Jev's evaluation pattern easy to see, run and reuse: Choice judgments, probability distributions, many questions in one call, custom Boolean, Choice or Score questions added from the UI, real-world state rather than synthetic prompts, and offline batch scoring against a frozen rubric. A committed static leaderboard scores many homepages with five Score sub-scores, for example 88 out of 100 for one site and 52 for another. A live demo and methodology document are provided. It is an unofficial project not affiliated with TypeSafe.

What the decision model decides

TypeQuestionAnswers
ChoiceWho is this product for, based on the homepage?—
ScoreHow clear is it what this product does?—
ChoiceWhat is the first change to make to this homepage?—

How it works

  1. Validate URL and fetch clean Markdown via ReplyNodes
  2. Ask Jev ten typed questions in one call
  3. Present inspectable findings and probabilities
  4. Batch-score many homepages for the leaderboard

Strengths

  • Output is explicit decisions, not a long generated review
  • Users can add their own questions
  • Static leaderboard dataset without a datastore

Caveats

  • Subjective marketing judgments
  • Unofficial project

At a glance

Type
Service
Status
beta
Language
TypeScript
Runtime
Web app (Vercel AI Gateway)
Has code
yes
License
Apache-2.0

Use cases and integrations

How curated lists describe it

  • Product evaluation: analyzes a public SaaS landing page as clean Markdown and asks Jev ten bounded `Choice` questions about first-visit understanding, returning inspectable findings for the first change to make.

    yibie/awesome-jev

  • Community project that analyzes a public SaaS landing page as clean Markdown and asks Jev ten bounded `Choice` questions about first-visit understanding, including the first change to make.

    AnotiaWang/awesome-decision-models

  • See what Jev thinks about your SaaS website — powered by ReplyNodes web context and Vercel AI Gateway.

    logicrw/awesome-jev-projects

  • Fetches a public SaaS page as Markdown, then asks Jev ten bounded questions about likely first-visit comprehension; URL validation, caching, sanitization, and presentation remain deterministic. [Demo](https://replynodes.com/jev-web-analyzer/).

    cobanov/awesome-jev

  • See what Jev thinks about your SaaS website — powered by ReplyNodes web context and Vercel AI Gateway.

    hellogumbo/awesome-jev

Frequently asked questions

What does it judge?

A homepage's audience, clarity, differentiation, call to action, trust and product motion, plus a first change to make.

Can I add questions?

Yes, Boolean, Choice or Score questions can be added from the UI.

Is it official?

No, it is an unofficial community project.

Related projects

  • jev-writer

    Analytics engine that rates published posts on pre-registered rubrics with Jev and tests ratings against real engagement.

    ProjectbetaTypeScript
  • pagegrade

    Chrome extension that grades readable page sections on ten rubric scores with Jev via Vercel AI Gateway and gives the page an A-E grade.

    ProjectbetaTypeScript
  • typesafe-eval

    Document evaluation CLI and CI gate using Jev or OpenAI Decisions, with pre-registered validation and documented reliability limits.

    CLIbetaPython
  • judging-with-typesafe

    Russian-language Letta agent skill that uses Jev as a fast external judge, with rules for treating mid-range probabilities as 'unsure'.

    ProjectprototypePython
  • jev-geo-audit

    Audit of 300 web pages for AI citability with nine Jev decisions each, checked against an LLM judge for agreement, cost and latency.

    ProjectresearchPython
  • clarity-judge

    Writing checker running separate named Jev checks such as hedging, filler, tone and passive voice, each with its own verdict.

    DemobetaTypeScript