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.

Rhetoric Audit is designed to be approachable for anyone who reads critically — from researchers and journalists to curious general readers. This page answers the questions users ask most often, organized by topic.

​
Getting started

Yes. You get 10 free analyses when you sign up — no credit card required. After those are used, your free signed-in tier gives you 50 scans per day, every day, at no cost. If you need more than 50 scans in a single day, you can top up with a credit pack.
No. The web dashboard at rhetoricaudit.com works fully on its own — you can paste article text directly and run an analysis without installing anything. The Chrome extension adds convenience by detecting when you’re on a news article page and letting you analyze it in one click, but it’s entirely optional.
Any news article, opinion piece, or editorial that contains at least 100 characters of text. For the best results, use articles that are 400 words or longer — the FME pipeline produces its most precise output on substantive news and opinion writing. Press releases, homepages, and search results are filtered out automatically.
Rhetoric Audit is primarily optimized for English-language content. You can submit text in other languages, but accuracy may be lower — the underlying FME models were benchmarked on English articles, and non-English content may produce less reliable scores across some parameters.

​
Analysis and results

Most analyses complete in 2–5 seconds. If the exact URL you submitted has been analyzed before, the result is returned instantly from the deduplication cache — no credit is deducted for cached results.
Rhetoric Audit achieved 100% benchmark band accuracy on the FME V19.1 Suite C evaluation, which tested 14 articles across 5 ideological strata. The hallucination rate is 0%. Every score traces back to exact quoted text in your article — there are no invented or fabricated claims in the output.
You can review the full public benchmark report at rhetoricaudit.com/test-results.
When you see a cached: true indicator, it means the same URL was analyzed before by another user (or by you previously). The result is returned instantly from the deduplication cache, and no credit is deducted from your account. The analysis itself is identical in quality to a fresh scan.
Not directly via URL — Rhetoric Audit cannot retrieve content behind a paywall. Instead, copy the full article text from your browser (after you’ve accessed it through your subscription) and paste it into the analysis page on the web dashboard.
The ambiguity flag appears when the article’s content spans multiple ideological positions simultaneously, making a single classification less certain. This is a transparency signal, not an error — it means the confidence in the top ideological frame is lower than usual, and you should treat that particular score with more interpretive caution.
Logic Fractures (parameter 04) are specific logical fallacies detected in the article, each accompanied by an exact quoted passage from the text and an impact analysis. Rhetoric Audit detects 24 recognized fallacy types — including Ad Hominem, Straw Man, False Dichotomy, Appeal to Fear, and Bandwagon, among others. You can expand each Logic Fracture in the results UI to read the full quote and reasoning.

​
Credits and pricing

Each standard article scan costs 1 credit ($0.30). Intelligence Briefs — which pull cross-platform signals from X, Reddit, YouTube, News, and the open web — cost 2 credits per brief. A PDF export is included with every Intelligence Brief.
No. Credits you purchase never expire. You can use them at whatever pace suits your research.
When your paid credit balance reaches zero, you automatically fall back to the free signed-in tier, which gives you 50 scans per day. You won’t lose access — analysis continues uninterrupted within the free daily limit.
For billing questions or refund requests, reach out to the Rhetoric Audit support team via the contact options on the website. The team reviews requests on a case-by-case basis.

​
Chrome extension

The Chrome extension supports Google Chrome only. It is built on Manifest V3, Chrome’s current secure extension standard. There is no Firefox, Safari, or Edge version at this time.
The extension needs broad URL permissions so it can detect when you’re currently viewing a news article page — regardless of which news site you’re on. Without this permission, the extension could not automatically recognize article pages or extract text for analysis. The extension does not read or transmit page content unless you explicitly click the Rhetoric Audit icon to trigger an analysis.
The extension can only analyze text that is visible and accessible in your browser. If you have a subscription to a paywalled publication and the article text is rendered on your screen, the extension will analyze that visible text. It cannot bypass paywalls or access content that isn’t displayed to you.

​
Privacy and data

Yes. When you submit an article, up to 10,000 characters of the article text are stored as part of the deduplication cache. This cache allows subsequent analyses of the same URL to return instantly and at no credit cost to any user. Article text is used solely to improve analysis quality and is not shared or sold.
By default, every completed scan generates a public shareable shortlink that anyone with the link can view. If you prefer your scans not to be publicly accessible, you can adjust your privacy settings in your account settings page.