New: Use Eli in ChatGPT and Claude

Free tool · No signup

Would AI use this page as a source?

Paste any public URL. We grade how useful the page is as an answer and recommendation source: direct answers, evidence, structure, freshness and technical access.

Usually 3–8 seconds. No AI calls, no signup.

What it checks

Five sub-scores and the five highest-impact fixes for that page.

01

Answerability

A clear subject, a direct answer near the top, question coverage and a clean heading hierarchy.

02

Citability and evidence

Statistics, examples, comparisons, first-party proof, external citations, author and dates.

03

Commercial intent and access

Product clarity, a reachable next step, structured data, semantic HTML, crawlability and canonicals.

When to use it

Example use cases

  • Before publishing a comparison or pricing page: check it can stand alone as an answer.
  • When a strong page is never cited: find whether evidence, freshness or access is the gap.
  • Triage a content backlog: grade the pages that matter commercially first.

Methodology

How the result is produced

Analysis version 1.0.0 · Scoring methodology 2026-10-04.1 · Updated 2026-10-04

  1. One fetch of the page, without running JavaScript. If the content needs JavaScript to appear, that is itself a finding.
  2. Roughly twenty-five fixed rules across five categories. Each rule has a weight and returns pass, warning, fail or not tested.
  3. No AI model assigns the score. The same HTML always produces the same result.
  4. Performance signals are limited to what one request reveals (response time, HTML size, script count). No lab performance test is run and none is implied.

If something cannot be tested, the report says “Not tested” or “Unable to verify”. Uncertainty is never turned into a score.

FAQ

Questions about this tool

Is a high score a guarantee that AI will cite the page?

No. The score measures properties that make a page usable as a source. Whether an assistant cites it also depends on the question, the competition and external authority.

Why is the score not generated by AI?

A model-generated score cannot be explained or reproduced. Every point here maps to a visible rule and its evidence.

The suite

Related free tools

All free tools

A free check finds the problem. Eli keeps fixing it.

Run the free check first. If the gaps matter, connect Eli and let it keep working on them.

Connect Eli