AI Content Audit

Free, no signup. One pass over your content gives you a single GEO/AEO score built from three lenses — is it written to be quoted, shaped to be answered from, and sourced well enough to be trusted — plus a merged, prioritized fix list.

Built by Patrick Stox

Before

A long article with no direct answer, unsupported statistics, and dense academic prose.

After

A page that opens with a direct answer, quotable sentences, and sourced claims — ready for AI retrieval and citation.

What each lens measures
Lens	Measures	Weight
Extractability	Readability grade and mean sentence quotability — can a passage stand alone when lifted?	35%
Structure	Answer-ready citability — direct answers, chunk relevance, freshness, author signal	40%
Trust	Citation coverage — unsupported claims, aging evidence, broken citations	25%

Or fetch a public URL

Pasted content is analyzed in your browser. A URL is sent only to the protected fetch endpoint; paste content takes priority. After a successful URL check, only its domain is remembered locally.

Paste article text or HTML

Checks run from our server; we fetch the URL you enter and don't keep the results. The three-lens audit runs in your browser. Up to 25 cited URLs may be sent to the bounded status checker; pasted prose is not sent. Anonymous run-level outcome counters may be used for aggregate research; URLs, domains, IPs, and identifiers are never included, and no statistic is released below 100 runs.

Feedback

Sample report Built-in deterministic example

The Try an example button loads a short, attributed paragraph with one aging statistic. The current engine returns:

GEO/AEO score: a numeric 0–100 result

Extractability · Structure · Trust — each shown as a separate sub-score

✓ Direct answer near the start
✓ Visible update date
✓ Author or reviewer signal

One prioritized fix: review the aging 2021 statistic

How to use it

Paste one page's visible text or HTML, or fetch a public URL.
Select Run content audit, or use the built-in example to see the output shape.
Start with the overall score and the three sub-score bars. Click a bar to expand its lens detail.
Work through the prioritized fix list, highest-impact first. Re-run after changes to confirm the score improves.

Scoring weights
Lens	Weight	Signal
Extractability	35%	Readability grade, mean sentence quotability
Structure	40%	Chunk relevance, answer coverage, quotability density, direct answer, freshness, author
Trust	25%	Citation gaps, aging evidence, broken citations

Local data

Saved targets, named lists, and recent check summaries remain only in this browser.

What the results mean

Overall GEO/AEO score — a weighted blend of the three sub-scores using the weights shown above. Higher is better, but the score is a heuristic, not a retrieval or citation prediction.
Extractability — how readable and quotable the content is. Penalized for high grade level and low mean sentence quotability.
Structure — how well the page gives an answer engine something clear to retrieve. Driven by chunk relevance, answer coverage, quotability density, and three content checks (direct answer, freshness, author).
Trust — citation and evidence quality. Penalized for unsupported claims, aging statistics, and broken or questionable citations.
Prioritized fix list — signals from all three lenses merged into one list, ranked by approximate score impact. Fixes that span two lenses (e.g. a buried answer that affects both extractability and structure) appear once with both tags.

How it works

The content extractor normalizes the input (HTML or plain text) into readable blocks. All three engines then run over the same extracted text: the readability scorer computes grade levels and sentence quotability; the answer-ready inspector checks chunk relevance, answer coverage, and three content patterns; the citation-gap analyzer finds unsupported claims and sends up to 25 cited URLs to the bounded status checker. Each lens normalizes its raw output into a 0–100 sub-score using the documented rules above, and the overall score is the weighted sum. The fix-list builder maps each lens signal to a fix entry, deduplicates overlapping signals, and ranks by impact.

Features

Single 0–100 GEO/AEO score with all three sub-scores visible.
Three expandable lens detail panels with the same depth as the standalone tools.
Merged, impact-ranked fix list deduplicated across lenses.
Text or HTML input; optional URL fetch via the shared /api/text-fetch endpoint.
Citation link-status pass reuses the bounded /api/check-status checker (up to 25 URLs).
CSV ticket export of all findings.
Everything runs in the browser; no model is called.

Limitations

The readability and quotability engines are English-oriented and can misread names, abbreviations, and specialized vocabulary. Answer-ready checks detect simple wording patterns, not meaning. Citation adjacency is a crude proxy: a URL in the sentence may be irrelevant, while a footnote may be valid but missed. The three-year aging threshold is intentionally generic. The bounded status checker does not read papers, assess source quality, or determine whether a citation entails the exact claim. The overall score is a transparent heuristic blend — it cannot predict whether any specific AI model will retrieve or cite the page.

Deep-dive tools

Each lens is also available as a focused standalone tool:

Quotable Sentence Finder — full readability and sentence-level quotability detail.
AI Citation Readiness Checker — citability with optional retrieval evidence and per-model freshness view.
Citation Gap Checker — full claim-level gap and aging-evidence report.

Frequently asked questions

Is this an AI visibility score?

No. The score is a transparent weighted blend of three deterministic writing heuristics: readability/quotability, answer-ready structure, and citation coverageAn AI citation is the visible source link an AI answer engine shows next to its generated text — the clickable reference that credits the web page it used. A citation's presence is a separate thing from whether the cited page actually supports the statement, and from being retrieved (read behind the scenes) or merely mentioned (named without a link); citation is driven more by brand mentions and being retrievable than by traditional ranking.. It does not observe or predict any specific AI model, answer engine, or ranking system.

Why are the weights 35/40/25?

Structure carries the largest share because a well-shaped answer passage is the single strongest citability signal in a content-only audit. Extractability (readability and quotability) is second. Trust (evidence and citation quality) is third because a well-structured, readable page can still be cited without perfect sourcing, but not without structure.

Does the tool check my links?

In URL mode the page sends up to 25 cited URLs to the bounded status checker. A failed status can flag a broken citation; the tool does not read page content behind the link or verify that a reachable source supports the claim.

Can a high score guarantee an AI citation?

No. A high score means the content satisfies the three local heuristics. Whether an answer engine retrieves and citesAnswer Engine Optimization (AEO) is the practice of structuring content so engines deliver it as a direct answer — featured snippets, voice assistants, and AI search — rather than just a ranked link. Coined for voice search in 2018 and revived for the LLM era. Google's position is that it's still SEO. a page depends on external factors (rank, authority, indexation, model behavior) that this tool deliberately does not measure.

Why is a claim flagged when the sentence links to a source?

The tool checks whether a URL appears in the same sentence as a claim. A footnote, endnote, or nearby link may be valid but still missed by the adjacency heuristic. Always verify the cited source supports the specific claim.

Is pasted content uploaded?

No. The three-lens audit runs entirely in your browser. Only the optional URL fetch and the bounded status check (up to 25 URLs) involve a server call.

Next stepQuotability & Entity-Preserving Rewriter — generate the corrected version.

Feature requests for Ai Content Audit

Upvote what you want most. New ideas can be submitted from the floating Feedback menu; requests appear here once approved, and the most-wanted rise to the top.

Loading…

➕ Request a feature

New requests are reviewed before they appear here.