Skip to main content

Documentation Index

Fetch the complete documentation index at: https://www.rhetoricaudit.com/docs/llms.txt

Use this file to discover all available pages before exploring further.

Most issues with Rhetoric Audit fall into a small number of predictable categories. Work through the relevant section below to resolve your problem quickly.
If you can’t find your answer here, reach out to the support team via the contact options on the Rhetoric Audit website.

​
Analysis issues

The analysis engine requires a minimum of 100 characters of article text to produce meaningful results.How to fix it:
  • If you’re using the web dashboard, paste more of the article text into the input field. Copy the full article body rather than just a headline or excerpt.
  • If you’re using the Chrome extension, this message typically means the extension extracted very little text from the page. Try copying the full article text manually and pasting it into the web dashboard instead.
This error means the junk filter rejected the content you submitted. The filter is designed to prevent analyses of pages that aren’t editorial articles — such as homepages, category pages, search results, or navigation-heavy pages.How to fix it:
  • Make sure you’re on a specific article page, not a site homepage, tag page, or search results page.
  • Navigate directly to the article’s own URL before triggering the analysis.
  • If you’re on a valid article page and still seeing this error, the page may use a non-standard layout. Try copying the article text and pasting it directly into the web dashboard.
You’ve used all your free scans for today.
  • Anonymous (not signed in): The limit is 10 scans per day.
  • Signed in (free tier): The limit is 50 scans per day.
Both limits reset at midnight UTC.How to fix it:
  • Wait until midnight UTC for the daily limit to reset.
  • Sign up for a free account if you’re currently analyzing anonymously — this immediately raises your limit to 50 scans/day.
  • Purchase a credit pack from your account dashboard to remove the daily cap for paid scans.
Cached results — analyses of URLs that have already been scanned before — do not count against your daily limit.
For a URL being analyzed for the first time, the full FME pipeline runs across multiple stages including span annotation, claim grounding, and cross-platform corroboration. This is normal and typically completes within 10–15 seconds under normal load.What to do:
  • Wait for the analysis to complete. Do not close the tab or navigate away.
  • If the analysis has been running for more than 30 seconds, refresh the page and try again.
  • For URLs that have been analyzed before, results return instantly from the cache.
A few factors can reduce the quality of results:
  • Short articles: The FME pipeline produces its most precise output on articles of 400+ words. Very short pieces provide insufficient signal for some parameters.
  • Non-English content: The pipeline is optimized for English. Other languages may produce lower accuracy across some scores.
  • Press releases and non-editorial content: The analysis is designed for news and opinion journalism. Press releases, corporate announcements, and marketing copy may produce unusual or less reliable scores.
  • Highly fragmented or list-based articles: Articles that consist mostly of bullet points or headers without full prose paragraphs may not yield strong narrative analysis.
If your article meets none of these conditions and results still seem off, try rerunning the analysis — a fresh LLM call will run if the cache doesn’t have a hit.

​
Chrome extension issues

By default, Chrome hides newly installed extensions in the extensions overflow menu.
1

Open the extensions menu

Click the puzzle-piece icon at the right end of your Chrome toolbar.
2

Find Rhetoric Audit

Locate “Rhetoric Audit” in the dropdown list.
3

Pin it to your toolbar

Click the pin icon next to Rhetoric Audit. The icon will now appear permanently in your toolbar.
Some news sites render their article content using JavaScript after the page loads, which can prevent the extension from extracting the text correctly.How to fix it:
  • Wait a moment after the page fully loads before clicking the extension icon.
  • If the error persists, copy the article text manually from the page and paste it into the web dashboard at rhetoricaudit.com to run the analysis there instead.
If clicking the extension icon does nothing or the extension appears grayed out:
  • Open Chrome’s extension management page (chrome://extensions) and confirm that Rhetoric Audit is enabled (the toggle should be blue).
  • If you’re on a corporate or enterprise-managed Chrome profile, your organization’s policy may restrict extension activity on certain sites. Contact your IT administrator or use the web dashboard at rhetoricaudit.com instead.
  • Try disabling and re-enabling the extension from the extensions management page.
Click the Rhetoric Audit extension icon in your toolbar. If you see a sign-in prompt, click it and authenticate with your Rhetoric Audit account credentials. Your session will then carry over and your scan history and credits will be accessible from the extension.

​
Account and billing

Verification emails occasionally land in spam or junk folders.
1

Check your spam folder

Look for an email from Rhetoric Audit in your spam or junk folder. Mark it as not spam if you find it there.
2

Resend the verification

Return to the sign-up success page and click the Resend verification link to request a new email.
3

Check your email address

Make sure you signed up with the correct email address. If you entered a typo, you’ll need to start the sign-up process again with the correct address.
Credit additions are usually instant but can occasionally take a moment to propagate.What to do:
  1. Wait 30 seconds after completing payment.
  2. Refresh the page or navigate back to your dashboard.
  3. If your credit balance still hasn’t updated after a minute, contact the Rhetoric Audit support team with your payment confirmation details.
Scan history works differently depending on whether you were signed in when you ran your analyses.
  • Signed-in users: Your full scan history is stored in your account and accessible from the dashboard history page on any device.
  • Anonymous users: History is tied to your browser and device, not to an account. If you analyzed articles while not signed in and then created an account on the same device, your past anonymous scans are automatically reattributed to your new account. If you signed up on a different device, the anonymous scan history on your original device cannot be recovered.
To ensure your scan history is always preserved across devices, sign in before running analyses.

​
API issues

The request is missing valid identity credentials. Every request to the analysis API must include either a userId (for signed-in users) or a deviceId (for extension and anonymous web requests). Check that your request body includes one of these fields and that the value is not empty or malformed.
This error indicates a temporary problem with the upstream AI model provider, not with Rhetoric Audit itself.What to do:Retry the request using exponential backoff — for example, wait 1 second before the first retry, 2 seconds before the second, 4 seconds before the third, and so on. The service typically recovers within a few minutes.
This is a rare error caused by the language model returning output that couldn’t be parsed into the expected format. It does not indicate a problem with your article or input.What to do:Retry the request once. The error is almost always transient. If it persists across multiple retries on the same article, contact support.