Skip to content
Free tool

Content freshness scanner

Paste one URL. A language model reads the page and lists the statistics, tool references, regulatory citations and time-bound phrases that have gone stale, in a suggested order of repair.

The analysis is model-generated. It is one model’s reading of one page at one moment, which makes it a starting point for a refresh rather than a finding. Caul is our measurement instrument and it is not involved here, so nothing this returns is a measurement, a share or a citation figure.

One public page. The scanner reads the HTML as served, so pages that build their content in the browser will look emptier than they are.
NO SIGN-UP · NO EMAIL · RATE LIMITED PER IP
Nothing has run yet

Enter a URL above. What comes back is one model’s reading of one page, produced in about a minute, and the limits below apply to every line of it.

What it checksfive categories, one page at a time

Content goes stale in a small number of predictable ways, and the prompt asks for exactly those. Anything genuinely evergreen is meant to come back unflagged, so a short list is a normal result rather than a failed scan.

Statistics and data claims
Figures attached to a year, market sizes, survey results and percentages whose stated date has run past the age thresholds in the prompt.
Technology and tool references
Tools, APIs, platforms and version numbers the model reads as deprecated, superseded or sunset.
Regulatory and compliance
Regulations, frameworks and legal requirements the model reads as having changed since the page was written.
Competitive landscape
Companies it reads as acquired, rebranded or closed, and competitor claims resting on old data.
Seasonal and time sensitive
Year-stamped predictions, past events, and relative phrasing such as recently, this year, new and upcoming.
How it scores

The P0 to P3 priority scale, and nothing else.

There is no composite, no grade and no number out of 100. Every flag gets one of four levels on the P0 to P3 scale below, and the tool reports how many landed at each. A count of flags is not a score of your page, and we do not convert it into one.

The scale is a rubric written into the prompt, not a calibrated instrument. The model applies it by judgement, the thresholds are ours rather than anyone’s standard, and two runs can place the same line differently.

P0 · Critical
A stat the model reads as wrong, a regulatory claim it reads as broken, or a deprecated tool presented as current.
FIX IMMEDIATELY
P1 · High
Statistics older than 18 months in prominent copy, or a year reference two or more years behind.
FIX THIS WEEK
P2 · Medium
Statistics older than 12 months, or a tool version slightly behind current.
FIX THIS MONTH
P3 · Low
Minor date references and relative time language that has started to drift.
NEXT REFRESH CYCLE
What it cannot seestated before you run it, not after

Six things this tool is blind to.

A free tool that overstates itself costs more credibility than it earns. These limits hold on every run, and knowing them is the difference between using the output and believing it.

Whether a figure is actually wrong.
It reads age and phrasing, not truth. A ten-year-old statistic that still holds gets flagged, and a fresh statistic that is wrong does not. Every flag needs checking against the source before anything on the page changes.
Anything the raw HTML does not contain.
One page is fetched once and parsed as served. Content assembled in the browser after load, content behind a login, and anything the server withholds from an automated request are all invisible to it.
Events after the model's training cutoff.
A tool deprecated last month, or a company acquired last month, can pass unflagged. The opposite also happens: something current gets flagged as stale because the model last saw an older state of it.
When the page was actually published.
Publication and modification dates are read from the page itself. A page that states no dates, or states dates that are not true, is assessed on its wording alone.
Whether any of this affects AI visibility.
This tool never queries an engine. It cannot tell you whether ChatGPT, Gemini or AI Overviews name you, cite you, or have ever read this page. That is a separate instrument and a separate question.
The same answer twice.
One page, one run, one model. There is no frozen prompt set behind it and no repeat runs, so a second scan can return a different list. Treat a single run as a reading rather than a result.
Method

The disclosure for a heuristic.

instrument
A general-purpose language model, called through the Vercel AI Gateway. Not Caul, and not a measurement instrument.
what was measured
Nothing is measured. A model is asked to read one page and flag content it judges to be stale, against a written rubric.
how
The page is fetched once as served, parsed to text, headings, structured data and FAQ blocks, and passed to the model with the rubric. The reply is checked for shape, not for accuracy.
over what window
The moment you press the button. One page, one fetch, one run, no repeat.
what this cannot tell you
It assesses age and phrasing, never truth, so a flag is a prompt to check rather than a defect found. It cannot read content the raw HTML does not carry, cannot know about changes after the model's training cutoff, and cannot tell you anything about whether AI answers name or cite you. Two runs on the same page can differ.

For the question this cannot answer.

Whether AI answers name you, and which competitor gets named instead, is measured rather than assessed: a fixed prompt set, run repeatedly across the engines, with coverage, run count and window attached to every figure. The audit is free, delivered on a call, and you keep the report and the underlying data whatever you decide.

What is in the reportAll free tools